An Evaluation of Factors Related to the Disproportionate
Representation of Children of Color in
Santa Clara County's Child Welfare System:
Child and Family Characteristics
and Pathways Through the System
Phase 2
Final Report

Submitted to the County of Santa Clara
Social Services Agency
Department of Family and Children's Services

By

The Child Welfare Research Team
College of Social Work
San Jose State University

August 31, 2002
(revised October 7, 2002)

Principal Contributors
Child Welfare Research Team

Alice M Hines, Ph.D.

Principal Investigator

Peter Allen Lee, Ph.D.

Co-Investigator and Primary
Data Analyst

Laurie Drabb/e, Ph.D.

Co-Investigator and Primary
Researcher on Key Informant
Interview Segment

Lonnie R. Snowden, Ph.D.

School of Social Welfare,U niversity of California at
Berkeley, Consultant on Data
Collection and Data Analysis

Kathy Lemon, MS. W.

Senior Research Assistant

TABLE OF CONTENTS
Page

EXECUTIVE SUMMARY

i

ACKNOWLEDGEMENTS

vii

I.

Introduction

1

II.

Study Objectives and Approach/or Phase 2

4

III.

Summary ofRelated Research and Relevant Santa Clara
County Statistics

7

Parent and Family-Related Factors and Child Maltreatment

7

Bias in Initial Reporting and Subsequent Service Delivery

12

Poverty and Characteristics of Impoverished Communities
That Increase the Risk of Children of Color Entering and Staying
In the Child Welfare System

14

The Impact of Recent CW Policy Initiatives on the Involvement
Of Children of Color in the Child Welfare System

18

Summary

22

Phase 2 Study Methods and Procedures

23

Case Record Review - Procedures

23

Case Record Review - Data Analysis

24

CWS/CMS Database of Closed Cases - Procedures
and Data Analysis

25

Key Informant Interviews - Procedures and Analysis

26

Pathways Through the Child Welfare System:
Case Studies by Ethnic Group

28

African American
Family Reunification Case
Family Maintenance Case

28
28
29

IV.

V.

TABLE OF CONTENTS
White
Family Reunification Case
. Family Maintenance Case

30
30
31

Hispanic/Latino
Family Reunification Case
Family Maintenance Case

32
32
34

Asian American/Pacific Islander
Family Reunification Case
Family Maintenance· Case

35
35

Overall Closed Case Sample: Results

VI.

37
39

Child Demographics and System-Related Characteristics
Child Characteristics
·
System-Related Factors

39
40
41

Summary of Key Findings from the Overall Closed Case Sample

51

Case Record Reviews

VII.

VIII.

Page

53

1.

General Child and Family Characteristics by Ethnic Group
Child Characteristics
Father Characteristics
Mother Characteristics

56
56
61
62

2.

System-Related Characteristics by Ethnic Group

73

3.

Services and Other Interventions Ordered at the
Jurisdictional/Dispositional Hearing by Ethnic Group

86

4.

Hearings and Case Changes by Ethnic Group

94

Key Informant Interviews
Key Informant Interviews- Methods

100

100

TABLE OF CONTENTS
Page

IX.

Key Informant Interview-Findings

IO I

Social and Economic Factors

101

Individual Bias Based on Race, Class, and Immigration Status

I 02

Systemic Inadequacies and Structural Bias

I 03

Factors Related to Family Characteristics

I 04

Factors Related to Laws and Policies

I 04

Interviewee Recommendations

I 05

Summary and Conclusions

108

Pathways Through the Child Welfare System: Summary Descriptions by
Ethnic Group and Implications for the Child Welfare System

I 08

African Americans
Child and Family Characteristics
System-Related Factors
Implications for Intervention with African American
Children and Families in the Child Welfare System

I 08
I 08
109
110

White
Child and Family Characteristics
System-Related Factors
Implications for Intervention with White
Children and Families in the Child Welfare System

111
111
111
112

Hispanic/Latino

112

Child and Family Characteristics
System-Related Factors
Implications for Intervention with Hispanic/Latino
Children and· Families in the Child Welfare System

112
113
114

TABLE OF CONTENTS
Page
Asian American/Pacific Islander
Child and Family Characteristics
System-Related Factors
Implications for Intervention with Asian American/
Pacific Islander Children and Families in the
Child Welfare System
Overall Conclusions

References

114
114
115

116
119
124

Attachments
1. Sampling Procedures
2. Data Collection Process and Procedures
3. Child Welfare System Case Record Data Extraction Form
4. Reliability Study
5 .. Variable Checklist
6. Sampling Summary: Case Availability Status Analysis

1. Table 1:

Demographic and System Characteristics of Children, Santa Clara
County Open Cases, December 2000.

Table 2:

Demographic Characteristics for Children in Out of Home
Placement (OHP), Santa Clara County Open Cases, December
2000.

Table 3:

Case Characteristics for Children in Out of Home Placement
(OHP), Santa Clara County Open Cases, December 2000

Table 4:

Demographic and System Characteristics by Race/Ethnicity in Out
of Home Placement (OHP}, Santa Clara County Open Cases,
December 2000

Executive Summary
Introduction
While child abuse and neglect appears to affect children of all racial and ethnic
origins (US Department of Health and Human Services, 1998; Sedlack & Broadhurst,
1996), an analysis of national, California and Santa Clara County data on the ethnicities
of children in out of home placement reveals that, compared to their presence in the
general population, there is a disproportionate involvement of children of color in the
public child welfare system (CWS). In Santa Clara County, in particular, when compared
to the general population, African American, Hispanic/Latino and Native American
children are overrepresented in the CWS, while Asian American/Pacific Islander and
White children are underrepresented. African Americans represent 4% of the general
child population in the county, and are 14.7% of children in the CWS. Hispanic/Latino
children represent 30% of the general child population in Santa Clara County and
constitute 53.5% of the child welfare cases. Native Americans are approximately 0.5% of
Santa Clara County's population and represent 1.0% of children in the CWS. Asian
American/Pacific Islander children represent 21 % of the general county child population
and 5.1 % of children in the CWS; Whites constitute 45% of the general child population
and 25.8% of the child welfare population (Needell et al., 2002, US Bureau of the
Census, 2000).
The disproportionate involvement of children of color in the CWS has long been
an issue of concern for CWS workers, clients, researchers and government and
community groups; yet no research to date (with the exception of this study) has
systematically investigated the factors associated with this disproportionality. In an effort
to understand better the factors related to the disproportionate number of children of color
in the CWS in Santa Clara County, the Department of Family and Children's Services
(DFCS) contracted with the Child Welfare Research Team (CWRT) in the College of
Social Work at San Jose State University to conduct a three-year study on this topic. An
advisory group, consisting of administrators and representatives from various
racial/ethnic groups was convened to help guide the development of the project.
The primary overall question posed by DFCS was: What are the primary reasons
why children of color are disproportionately represented in Santa Clara County's Child
Welfare System? In order to address the complexity of this question, the CWRT elected
to employ a multiphase/multimethod approach, beginning with an initial exploratory
phase that was completed in April 2001.
This report provides findings from Phase 2 that ran from September 2001 to
August 31, 2002. Specifically, in this second phase of the study, the Child Welfare
Research Team (CWRT) addressed two of the four themes that emerged from Phase 1
(see Section II of the current report for a review of all four themes presented at the end of
Phase 1). The overarching themes guiding the current Phase 2 report are as follows: l)
little is known about specific pathways through the CWS and ways in which these

pathways differ for various racial/ethnic groups, and 2) various racial/ethnic groups may
receive different treatment at key decision making points in the system.
The primary methodology for Phase 2 involved extensive, in-depth reviews of
403 closed child welfare case records, a parallel descriptive analysis of 1720 closed cases
within the CWS/CMS database, and key informant interviews with managers and
supervisors in the county's Department of Family and Children Services (DFCS).

Overall Conclusions
Combining results from our overall sample of 1720 closed cases, the in-depth
record review of 403 child welfare cases and interviews with 8 key informants in
managerial positions in DFCS, the following overall conclusions can be drawn:

1.

Descriptive narratives presented in Section IX of this report indicate that the
families belonging to each of the four racial/ethnic groups present unique and
diverse profiles.
•

African American families are largely headed by young, single, isolated, poor
mothers who suffer from substance abuse and have experienced criminal
involvement and domestic violence. At the same time, their reported rate of
mental health problems is low.

•

White families are made up of mothers who are also single, but who are better
educated than average and who are less likely to be on welfare. While
substance abuse, mental health problems and domestic violence are prevalent
within this group, white mothers tend to have a lower than average rate of
involvement in the criminal justice system.

•

Latino families are predominantly poor, with a higher than average number of
children. Many Latina mothers (18.4%) are non-English speaking and
approximately 26% are foreign born. Latina mothers are more often living
without a spouse, have low levels of education and experience problems
including substance abuse, and criminal activity, as well as domestic violence.
Like African American mothers, their reported rates of mental health
problems are low. For the immigrant Latino families, issues related to
immigration, legal status and the ability to secure employment may also have
an impact on their welfare, but were undetectable with the available data.

•

Asian/Pacific Islander families who are assigned to involuntary child welfare
services, are also often non-English speaking and most likely to be foreign
born. Most Asian/Pacific Islander mothers in this group are older, have fewer
children than average, have less education than average and are married and
living with a spouse. Mental health problems and involvement in the mental
health system rather than substance abuse or criminality characterize this
group. This finding is notable as available research indicates that elevated

ii

rates of mental health problems are uncharacteristic of immigrant populations.
In addition, rates of physical abuse are higher than average for this group and
Asian American/Pacific Islander fathers are comparatively more often
reported as the perpetrators of abuse.
2. Families within each ofthe four racial/ethnic groups are extremely vulnerable.
•

The characteristics that emerged in our narrative descriptions as key
characteristics of each of the four racial/ethnic groups have been associated in
the research literature as risk factors for child abuse and neglect. These
characteristics have also been associated with increased serious behavioral
and adjustment problems in children and adolescents.

•

Comments by key informants underscored the relationship between
characteristics including poverty, lack of education, insufficient job skills, as
well as involvement with drugs and violence, and bias on the part of workers,
as contributing to the overrepresentation of African American and Latino
children in the CWS.

•

While Asian American/Pacific Islander families have traditionally been
underrepresented in the CWS, our results suggest that Asian American/Pacific
Islander families with certain characteristics are emerging as a high-risk
group. More information and careful monitoring of this group is clearly
warranted.

•

Identifying vulnerable families is an important piece of the puzzle in
explaining involvement and retention in the CWS. Understanding family
characteristics that represent risk factors can help systems design interventions
that offset the potential for CWS involvement and prolonged involvement.

3. Once in the Child Welfare System, children in each ofthe four racial/ethnic groups

follow different pathways and experience different outcomes.
•

African American children are younger than average when they enter the system,
have higher than average rates of being assigned to family reunification services
and are initially placed with a relative. African American children experience
more court hearings, have a higher than average rate of being removed from their
families, longer than average stays in each out-of-home placement and a longer
average total case duration than children in other groups. At case closure, African
American children are less likely to be reunified with their families and most
frequently in permanent placement. ·

•

White children are older than average on entry to the CWS and though they are
also most often assigned to family reunification services and placed with a
relative, they tend to have a shorter than average stay in each out of home
placement and a lower than average total case length. At case closure, White
children are most likely to be in family maintenance services.

iii

•

Likewise, Latino children are also most often assigned to family reunification
services and placed with a relative at initial placement. They experience shorter
than average stays in each out-of-home placement, have a shorter total length of
time in out-of-home placement and a shorter total case length. Results also
indicate that Latino children experience a relatively high number of unique
placement homes, suggesting that they experience multiple transitions as they
wend their way through the system. For Latino children, their most common
status at case closure is permanent placement, suggesting that many Latino
children are not reunified with their family of origin. Similar to African American
families, they experience a lower than average rate of family maintenance
services.

•

Asian American/Pacific Islander children also tend to be assigned to family
reunification services, but receive family maintenance services at a higher rate
than other groups. Rather than being placed with relatives, their initial placement
is likely to be in a family foster home. Asian American/Pacific Islander children
appear to have lengthier than average stays in each out-of-home placement and
longer than average total time in out-of-home placement, as well as a longer than
average total case length. The most frequently occurring final out-of-home
placement for Asian American/Pacific Islander children tends to be a family
foster home, and they are more likely to have their case closed with permanent
placement services.

•

Once in the CWS, African American and Latino children tend not to return to
their families. This finding was underscored by focus group results in Phase 1 of
our study, as well as by comments from key informants during the current phase.
Key informants stated that worker bias, poverty, as well as immigration status
were key barriers to exiting the system. In addition, key informants suggested
that individual bias on the part of workers might be instrumental at both the front
end of the system and in decisions that are made once the child enters the system,
in prolonging stays of children of color.

4. The services ordered for families ofcolor are generally limited to a one-size-fits-all

approach and to a small array of available senlices. The reliance on traditional
formal services does not appear to meet the needs of these highly diverse
ethnic/racial family groups.
•

Parenting education, substance abuse treatment and counseling (both individual
and group) were the most commonly ordered services at the
jurisdictional/dispositional hearing. These results suggest that workers are
offering the same services for all families, regardless of their needs and rely
heavily on traditional, formal services.

•

Results also indicate that services are not distributed uniformly across
racial/ethnic groups and do not necessarily match their specific needs. In spite of
high rates of substance abuse problems and criminal involvement, African
American parents were most often ordered to receive parent education only.
iv

White families appear to receive more of the available services including
parenting education services, individual and group counseling and substance
abuse treatment services. In addition to parenting education services, Latino
families receive slightly higher than average rates of substance abuse services.
However, it is not clear whether these provided services are culturally and
linguistically appropriate. Neither African American nor Latino families appear
to receive mental health services. Asian American families appear to receive
mental health services at a higher rate than other groups, but again it is not clear
whether they are meeting the unique needs of this culturally distinct group.
•

The paucity of services ordered indicates that children and families of color are
not provided with sufficient preventive and supportive services and that
traditional CWS may not meet the particular needs of these unique and diverse
groups. These findings also corroborate those of a national forum on children and
families of color in the CWS (CWLA, 2002). In addition, research has indicated
that individual, group and couples counseling, as well as parenting training have
shown only minimal success in teaching parents better skills and reducing the
likelihood of further abuse in families marked by serious and chronic abuse
(Albee & Gullota, 1997; McLoyd, 1998).

•

There is a need for preventive and early intervention services for vulnerable
families of color. For example, home visitation services may be exceptionally
helpful in addressing the needs of vulnerable, at-risk families from diverse
racial/ethnic groups. Research has indicated that programs of home visitation that
promote positive health-related behaviors in mothers of young children,
competent care of their children and linkage with needed health care and human
services, reduce rates of criminality, problems related to substance abuse and
child abuse and neglect among young, unmarried, isolated poor mothers (Olds et
al., 1997, 1998). Studies that follow children of mothers involved in such home
visitation programs, into adolescence have found these youth have fewer serious
behavioral and adjustment problems than youth whose mothers did not participate
in home visitation programs. Indeed, the U.S. Advisory Board on Child Abuse
and Neglect has recommended that home-visitation services be made available to
all parents of young children as a means of preventing child abuse and neglect
(U.S. Department of Health and Human Services, Administration for Children
and Families, 1991).

•

The use of more non-traditional, culturally sensitive services that are conducted in
the client's primary Language are clearly -warranted. Key informants underscored
this point. A paucity of social services, particularly multi-lingual services, was
cited as a significant barrier for many families of color. Interviewees discussed
the shortage of substance abuse treatment programs, particularly those geared for
women with children and people whose primary language is not English. Multilingual and culturally appropriate domestic violence services, parenting classes,
and other social services were considered in need of development.

V

•

The form of mandated services is also an important issue. As one key informant
noted, "if a Latino parent is court ordered to therapy, this may be seen as 'being
crazy, while attending a psychoeducational group is (viewed as) less of a
problem."

•

The scarcity of accessible services was perceived as "discrimination against the
poor who have to rely on free treatment" by more than one informant.

•

In addition, key informants noted that recognition of the extended family system
is critical, particularly in relation to working with many communities of color.
"Thinking of family as a mother-father-child configuration is common but
problematic .. .it is just not how children are really raised."

•

More research and evaluation of interventions and programs targeted to children
and families of color is necessary. Rather then implementing untested
interventions, it would be more efficacious to begin with programs that have been
tested, replicated and found to work with families from unique and diverse
racial/ethnic groups.

5. There is a need to involve multiple social service systems in a comprehensive and
coordinated effort to meet the needs of children and families ofcolor.
•

Results from Phase 2 indicate that the problems experienced by families across
the different racial/ethnic groups span multiple systems including: mental health,
juvenile justice, adult criminal justice, substance abuse, and welfare.

•

Statistics presented in the literature review section of this report also indicate that
families of color are involved in systems other than child welfare in high
numbers. Prevention and intervention efforts should involve a deliberate and
organized coordination of these multiple systems.

vi

Acknowledgements
This report could not have been completed without the help of a very large
number of people.
We wish to thank the former Director of Santa Clara County Social Services
Agency, Yolanda Rinaldo, the initiator of an important discussion with Dr. Sylvia R.
Andrew, the Dean of the College of Social Work at Sari Jose State University, about the
possibility of a joint research project with .the Agency and the College.
Thank you to Leroy Martin, former Director of the Department ofFamily and
Children's Services, who chaired the Advisory Committee and provided valuable
oversight for the research project until his retirement and to Norma Doctor Sparks,
current Director of the Department of Family and Children's Services.
We extend a special appreciation to the members of the Santa Clara County
Board of Supervisors. This research could not have been conducted without ongoing
support from the Board of Supervisors, and we thank them for their consideration of this
very important issue.
Thank you to the Latino Social Work Network of Santa Clara County for their
continuing interest and support.
We truly appreciate the support and direction from our Advisory Committee
members from Social Services Agency: Cindy Amador, Jorge Gonzalez, and Rebecca
Rojas. Special thanks to Community members Noemi Baiza and Faith Fauntleroy who
provided key insights from their experiences as former employees and community
leaders. Thanks also to Mary Helen Doherty, formerly ofDFCS and current Executive
Director of Child Advocates.
Judi Boring and her staff, David Heinen and Yvonne Zing, were especially
helpful with providing access to the CWS/CMS system and arranging for the data set that
was used in this report. Judy Bushey was extremely helpful in guiding us through the
court report sections of the child welfare case records. Thank you to Stan Lee and his
staff who provided valuable training on the court report section of the child welfare case
records and to Maria Gonzalez who made the case records available for coding.
Our special thanks to Dean Sylvia R. Andrew who provided members of the
research team with the precious resource of matched release time.
The completion of Phase 2 was also made possible by our talented and dedicated
student research assistants: Jessica Coon, Susan Grace, Kathleen Kaminitz, Mandana
Mahdavi, Molly McNelley, Jennifer Morris, Tanesia Paige, Penny Sandhu, Elise
Travertini, and Bernard Voon. Their work on the case record reviews and data entry was
truly invaluable.

vii

Thanks to Shaaron Gilson for her assistance as Project Manager and for her
thoughtful comments on an initial draft of the report and to Yifei Ma for his consultation
on the construction of our longitudinal database.
In addition, this research could not have been carried out without the cooperation
of the staff of the agency, and we would like to acknowledge the importance of the
contributions of everyone who took the time to attend a key informant interview in order
to share ideas and perspectives with us.
Thank you to the staff at the San Jose State Foundation for their valuable
assistance in administering this important project.
Finally, a special thank you to Sharon lkami for her assistance in the production
of the final report.

viii

I.

Introduction

While child abuse and neglect appears to affect children of all racial and ethnic
origins (US Department of Health and Human Services, 1998; Sedlack & Broadhurst,
1996), an analysis of national, California and Santa Clara County data on the ethnicities
of children in out of home placement reveals that, compared to their presence in the
general population, there is a disproportionate involvement of children of color in the
public child welfare system (CWS). Recent national statistics indicate that African
American and Native American children are overrepresented in the CWS. African
American children represent 15% of the general child population, yet comprise
approximately 4 7% of children in the CWS and Native American children constitute
approximately 1% of the child population and represent approximately 2% of the CWS.
Though not all states provide data on Hispanic/Latino children in the CWS, aggregate
statistics from those that do reveal that while Hispanic/Latino children make up
approximately 16% of the national child population, they comprise 7% of the CWS
population. Asian American/Pacific Islanders and Whites also tend to be
underrepresented at the national level. Asian American/Pacific Islander children comprise
approximately 4% of the general child population and 1% of the CWS, while White
children who are approximately 64% of the general child population, constitute 36% of
children in the CWS (Child Welfare League of America, 1998; Department of Health and
Human Services, 1998; Federal Interagency Forum on Child and Family Statistics, 2001;
U.S. Bureau of the Census, 2001).
Further analysis at the California level provides important information on
differences and trends that may go undetected at the national level. In California, a large
and ethnically diverse state, African Americans constitute 6% of the general population
(32% of whom are below the age of 18), but represent approximately 36% of children in
the CWS. Hispanics/Latinos comprise approximately 32% of the general population in
California with 43% being below the age of 18, and constitute 32% of children in the
state CWS. Native Americans represent 0.5% of California's population and constitute
approximately 1.5% of the children in the CWS. Asian American/Pacific Islanders are
under-represented in California's CWS, as they constitute approximately 11 % of
California's general population, with approximately 30% being younger than 18, but
represent just 2% of the children in the CWS. While Whites comprise approximately 47%
of the population in the state, with only 20% being below the age of 18, they constitute
30% of the children in California's CWS (Public Policy Institute of California, 2001;
Needell, Webster, Cuccaro-Alamin, Armijo, Lee, Brookhart, Lery, Shaw & Kim, 2001;
U.S. Bureau of the Census, 2000).
In Santa Clara County, there also exists a disproportionate representation of
children of color in the County's CWS. When compared to the general population,
African American, Hispanic/Latino and Native American children are overrepresented in
Santa Clara County's CWS, while Asian American/Pacific Islander and White children
are underrepresented. African Americans represent 4% of the general child population in
the county, and are 14.7% of children in the CWS. Hispanic/Latino children represent
30% of the general child population in Santa Clara County and constitute 53.5% of the

2

child welfare cases. Native Americans are approximately 0.5% of Santa Clara County's
population and represent 1.0% of children in the CWS. Asian American/Pacific Islander
children represent 21 % of the general county child population and 5.1 % of children in the
CWS; Whites constitute 45% of the general child population and 25.8% of the child
welfare population (Needell et al., 2002, US Bureau of the Census, 2000).

Project Description
The disproportionate involvement of children of color in the CWS has long been
an issue of concern for CWS workers, clients, researchers and government and
community groups; yet no research to date (with the exception of this study) has
systematically investigated the factors associated with this disproportionality. In an effort
to better understand the factors related to the disproportionate number of children of color
in the CWS in Santa Clara County, the Department of Family and Children Services
(DFCS) contracted with the Child Welfare Research Team (CWRT) in the College of
Social Work at San Jose State University to conduct a three-year study on this topic. An
advisory group, consisting of administrators and representatives from various
racial/ethnic groups was convened to help guide the development of the project.
The primary overall question posed by DFCS was: What are the primary reasons
why children of color are disproportionately represented in Santa Clara County's Child
Welfare System? In order to address the complexity of this question, the CWRT elected
to employ a multiphase/multimethod approach, beginning with an initial exploratory
phase that was completed in April 2001.
This report provides· findings from Phase 2 that ran from September 2001 to
August 31, 2002. Specifically, in this second phase of the study, the Child Welfare
Research Team (CWRT) addressed two of the four themes that emerged from Phase I
(see Section II for a review of all four themes presented at the end of Phase 1). The
overarching themes guiding the current Phase 2 report are as follows: 1) little is kriow
about specific pathways through the CWS and ways in which these pathways differ for
various racial/ethnic groups, and 2) various racial/ethnic groups may receive different
treatment at key decision making points in the system.
The primary methodology for Phase 2 involved extensive, in-depth reviews of
403 closed child welfare case records, a parallel descriptive analysis of 1720 closed cases
within the CWS/CMS database, and key informant interviews with managers and
supervisors in the county's Department of Family and Children Services (DFCS).

Overview of the Report
The next section of this report provides background information on the four
themes from Phase 1 and the specific Phase 2 research questions that were identified
from these themes. The third section includes a summary of previous research findings
concerning the factors associated with the disproportionate representation of children of
color in the CWS. When possible, key statistics on the characteristics of children,

3

families and communities in Santa Clara County and related Phase I findings are also
included in this section in order to provide a contextual framework for the research
review and the findings presented later in this report. Section four describes Phase 2
study methods and procedures. Eight individual case studies are presented in section five,
followed by a description of the overall closed case sample of I 720 cases in section six.
Section seven provides findings from case record reviews of 403 closed child welfare
cases, followed by section eightwhich presents results of the key informant interviews.
Summary narrative descriptions of ethnic differences in pathways through the CWS and
overall conclusions and implications are presented in section nine of the report.
Definition of Terms Used in this Report

Racial/ethnic group refers to cultural heritage and country of origin. Racial/ethnic group
is used interchangeably with the terms, children of color and ethnic group.

Disproportionate representation ofchildren ofa particular racial/ethnic group refers to

the difference between the proportion of children of a particular racial or ethnic group in
the CWS and the proportion of children of a particular racial or ethnic group in the
general population.

Overrepresentation of children ofa particular racial/ethnic group exists when the
proportion of children of a certain racial or ethnic group in the CWS exceeds its
proportion in the general population.
Underrepresentation ofchildren ofa particular racial/ethnic group exists when the
proportion of children of a certain racial or ethnic group in the CWS is less than its
proportion in the general population.
African American refers to children of African American heritage. The term Black is
included in this category.

Asian American/Pacific Islander refers to children of Asian American and Pacific
Islander heritage. The following population groups are included: Asian Indian,
Cambodian, Chinese, Filipino, Guamanian, Hawaiian, Hmong, Japanese, Korean,
Laotian, Polynesian, Samoan and Vietnamese and other Asian American/Pacific
Islanders.

Hispanic/Latino/a refers to children of Mexican American and Latin American heritage.

The terms Hispanic and Latino/a are included in this category.

Native American refers to children of Native American heritage and includes those

designated as Alaska Natives/Aleuts. The term American Indian is included in this
category.

White refers to children of European heritage. The term Caucasian is included in this
category.

4

II.

Study Objectives and Approach for Phase 2

Phase 1 emerging themes and research guestions identified for Phase 2
Phase 1, which was concluded in March 2001, presented preliminary findings
from an exploratory investigation of the primary research question. Data for Phase 1 were
gathered from three sources: research literature at the national, state and county levels;
Santa Clara County's management information system (CWS/CMS); and focus group
discussions with professionals in Santa Clara County who provide child welfare services
as well as parents, caregivers and youth who are recipients ofCWS services. Four central
themes emerged from Phase 1 findings (please see the. final Phase 1 report for the
complete findings from which these themes were generated). The four emerging themes
from Phase 1 included:
Theme 1: Little is known about specific pathways through the CWS and ways in
which these pathways differ for various racial/ethnic groups.
Much of the research on children in the CWS focuses on factors related to
movement in and out of the system. The specific pathways through the system for
children of various ethnicities is largely unknown, and may provide important insights
into the factors related to the disproportionate involvement of children of color. Focus
groups participants in Phase 1 frequently stated that once a child of color enters the
system, it is very hard for that child to exit. Conclusions from Phase 1 indicated that

more research is needed that focuses on the actual experiences ofchildren in care and
the individual and family-related characteristics that are associated with these
experiences.
Phase 2 of this study addresses this theme. Specific research questions related to
Theme 1 include:

•
•
•
•

Do pathways through the system differ for different racial/ethnic groups?
What individual, family and system-relatedfactors are associated with varying
pathways?
Do outcomes differ for children following particular pathways?
At what points along these pathways do and should interventions occur?

Conclusions from Phase 1 also indicated that CWS/CMS data alone would not
provide the information needed to address these research questions. Thus, the
methodology for Phase 2 included extensive case record reviews on a sample large
enough to collect information on the individual child and family as well as system related
characteristics not included in the CWS/CMS database. In addition, review of a large
sample ofclosed cases allowed us to examine the entire histories of children as they
progressed through the CWS and to conduct analyses that were not possible with the
CWS/CMS cross-sectional dataset that we used in Phase 1.

5

Theme 2: Various racial/ethnic groups may receive different treatment at key
decision making points in the system.
Differential treatment of children of color in both the initial reporting of child
maltreatment and throughout the various stages within the CWS may exist. CWS/CMS
data for Santa Clara County corroborate the possible differential treatment of children of
color at various points in the system. Phase 1 analysis revealed that African American
children spent significantly more time in placement than their White, Hispanic/Latino,
and Asian American/Pacific Islander peers. Differences in placement type were also
found-although kincare (placement with a relative) was the most frequent out of home
placement (OHP) type across ethnicities, the second most frequent OHP type varied ·
among racial/ethnic groups. For African American, Native American, White, and
Hispanic/Latino children, the second most common OHP type was a Foster Family
Agency. For Asian American/Pacific Islander children, the second most frequent
placement was a Foster Family Home. Asian American/Pacific Islander (18%) children
were also placed at the Children's Shelter at higher percentages than children of other
racial/ethnic groups.
In addition, focus group participants in Phase 1 expressed concern for possible
racial/ethnic differences in ways in which services and resources were allocated and ways
in which the judicial system handled cases.
Phase 2 of the study addresses Theme 2. Specific research questions guiding this
inquiry include:

•
•

What are the critical transitions/(choice points) in the CWS?
What is the extent ofover/under-representation ofchildren of differing ethnicities
at major transitions/(choice points) in the system?
• Are there racial/ethnic differences as children progress through the system?
• Do children from differing racial/ethnic groups receive the same or different
services and resources as a result of their involvement in the CWS?
Case record review data were collected to provide extensive information on decisions
at key points in the system and progress through the system, as well as services ordered
for children at the time of case disposition and at subsequent hearings. Key informant
interviews with managers and supervisors in sections and departments at key decision
points in the system provided additional information pertaining to cultural and
environmental aspects of the agency.
Theme 3: System level changes recently initiated at the federal, state and local
level will undoubtedly have an impact on children of color in the CWS.
Recent federal policies, including the Multi-Ethnic Placement Act-Interethnic
Adoption Provision (MEPA-IEP, 1996); the Adoption and Safe Families Act (AFSA,
1997); the Personal Responsibility and Work Opportunity Reconciliation Act
(PR WORA, 1996); as well as state and local county policies all impact the ways in which

6

families and children of color enter and stay in the CWS. Recent policy shifts toward
expedited permanent placements for children in out-of-home care and shortened timelines
for reunification, as well as new restrictions for Temporary Assistance to Needy Families
(TANF) recipients (more than half of children in foster care are from welfare eligible
families) may increase the likelihood of children of color entering and staying in the
CWS-although much more research is still needed to determine this association.
Theme 4: Factors related to the disproportionate representation of children of
color in the CWS are multiple and complex.
No clear consensus from the research literature exists on how families and
children of color become and stay involved in the CWS. Available evidence suggests
that, rather than one primary cause, there appear to be numerous and interrelated factors
associated with disproportionate rates of children of color in the CWS. Factors found to
be associated with CWS involvement for children of all ethnicities, include but may not
be limited to: parental mental illness, substance abuse, and incarceration; spousal abuse;
living in poverty; living in neighborhoods with concentrated poverty, and living in
communities with low levels of social organization. For children of color specifically,
research also indicates that race and class biases in initial reporting and subsequent CWS
service delivery do exist.
The examination of Themes 1 and 2 has helped to bring into focus a clearer picture of
the experiences of children of various ethnicities in Santa Clara County's CWS-as well
as furthered our understanding of the factors related to the disproportionate representation
of children of color in the CWS. In order to provide additional background in which to
frame the results of Phase 2 (presented in sections V-VIII), section three presents a
summary of the research relevant to the disproportionate involvement of children of color
in the CWS, augmented by findings from Phase 1, as well as relevant statistics that
pertain to children and families in Santa Clara county.

7

III. Summary of Related Research and Santa Clara County
Statistics

Rather than one primary cause, there appear to be numerous interrelated factors
associated with the disproportionate representation of children of color in the CWS
(please see the final Phase I report for a complete review of the relevant research
literature on this topic). This section summarizes the relevant research literature by
addressing four general areas associated with CWS involvement: I) parent and family
related risk factors, 2) race and class biases in the initial reporting and subsequent
processing of children in the CWS, 3) social factors related to poverty, neighborhood
effects, and other community-level predictors, and 4) the possible impact of recent child
welfare policy initiatives on children of color. Related data on the characteristics of
children, families and communities in Santa Clara County, as well as pertinent findings
from Phase 1, are also presented in order to provide a contextual framework for ways in
which these four areas of interest may operate within this particular locality and to help
frame the findings presented later in this report.
Parent and Family-Related Factors and Child Maltreatment
The impact of parent and family-related risk factors on child welfare caseloads
has received considerable attention in the research literature. Most notably, risk factors
associated with child maltreatment or entrance into the CWS include parental substance
abuse, parental mental illness, domestic violence and parental incarceration.

Parental Substance Abuse and Child Welfare System Involvement

It is estimated that one-third to two-thirds of substantiated child maltreatment

reports involve parents that abuse substances-including both alcohol and/or illicit drugs
(U.S. Department of Health and Human Services, 1999). Data from the National Institute
for Mental Health's Epidemiologic Catchment Area survey indicates that a child with a
substance abusing parent is nearly three times more likely to suffer from maltreatment
than other children (Chaffin, Kelleher, & Hollenberg 1996). Substance-abusing families
are also more likely to neglect their children than abuse them; have their children enter
the CWS at a younger age; have children placed in foster care; and experience longer
stays in foster placement compared to other children (Bays, 1990; U.S. Department of
Health and Human Services, 1999; Walker, Zangrillo & Smith, 1991 ).
Research on families in the CWS indicates that substance-abusing families are
more likely to be White (4 7%) or African American (4 7%) than Hispanic/Latino ( 6%)
and they tend to have more overall problems than other families in the system (U.S.
Department of Health and Human Services, 1999; Semiedi, Radel & Nolan, 2001). For
instance, one study found that substance abusing African American families in the CWS
were twice as likely as non-substance abusers to suffer from myriad social problems
including poverty, dependence on welfare/AFDC (85%), single parent status (53%),

8

having a mother that did.not graduate from high school (67%), and living in substandard
housing or having housing difficulties (44%) (Walker, Zangrillo, & Smith, 1991).
Prenatal substance abuse has also been linked to involvement in the CWS, partly
because in many states a positive toxicology screen upon delivery is enough to warrant a
child welfare investigation. It is estimated that 200,000 to 750,000 infants born each year
have been exposed to one or more illicit drugs before birth (National Institute on Drug
Abuse, 1994; Vega, Kolody, Hwang & Noble, 1993). Rates of prenatal drug use vary
depending on methodology and samples utilized. On a national level, more White women
than African American women or Latinas self-report the use of any illicit drugs during
pregnancy, but African American women are more likely to self-report cocaine use
during pregnancy than are White women or Latinas (U.S. Department of Health and
Human Services, 1999). In a separate study of women delivering in California hospitals,
11 % of new mothers tested positive for any substance use during pregnancy (Vega, et al.,
1993). African American women were found to have the highest prevalence of total
substance use, White women had the second highest prevalence of one or more drugs and
Latinas yielded the second highest prevalence for alcohol use. Latinas accounted for 45%
of all women who tested positive for alcohol in the study. Asian women generally had a
lower prevalence for substance use than any other group.
While prenatal exposure to alcohol and illicit drugs has been linked to negative
developmental outcomes for children (Bays, 1990), there is a lack of reliable empirical
evidence linking prenatal substance abuse with subsequent child abuse and neglect
(Jaudes, Ekwo & Voorhis, 1995). Similarly, although parental substance abuse is
prevalent .in CWS populations, there is a lack of strong empirical evidence concerning the
relationship between parental substance abuse and child maltreatment. Reliable findings
that causally link parental substance abuse and child maltreatment are extremely limited.
Definitive findings linking substance abuse, child maltreatment and ethnicity are likewise
absent from the literature. Of the studies that have included ethnicity-most neglected to
include groups other than Whites and African Americans.

Parental Mental Illness and Child Welfare System Involvement
There is considerable evidence to suggest that parents in the CWS have an
increased likelihood of suffering from mental health problems. In studies using samples
of families who are in the CWS, the incidence of parental mental illness is significantly
greater than in matched comparison groups of parents not involved in the CWS. Rates of
particular diagnoses among maltreating parents vary depending on methodology and
samples utilized. Mood disorders have been found in 28% to 41 % of maltreating parents;
post traumatic stress disorder has been found in 9% to 43% of maltreating parents, and
alcoholism or other substance abuse disorders have been found in 32% to 43% of
maltreating parents (Bellis, Broussard, Herring, Wexler, Moritz & Benitez, 2001;
Famularo, Barnum, & Stone, 1986; Famulare, Kinscherff, & Fe_nton 1992).
Approximately 65% of maltreating mothers have be~n found to meet criteria for a
personality disorder and 26% for an anxiety disorder (Famulare, Kinscherff, & Fenton,
1992). Severe mental illness appears to be less common than other diagnoses among

9

maltreating parents; rates for schizophrenia and bipolar disorder vary from 1% to 4%
(Bellis, Broussard, Herring, Wexler, Moritz & Benitez, 2001; Famularo, Barnum, &
Stone, 1986).
In studies utilizing community samples of families not involved in the CWS, the
presence of parental mental illness has also been linked to self-reports of child
maltreatment. Dinwiddie and Bucholz (1993) utilized two separate community samples,
including the 1984 Epidemiologic Catchment Area (ECA) Study and found that
participants who self-reported child maltreatment were also more likely than non-abusers
to have a DSM-III diagnosis of alcohol abuse/dependence, major depression and/or antisocial personality disorder, after the effects of age, gender and socioeconomic status were
controlled. In a separate analysis of 1984 ECA data, a lifetime history of any mental
disorder was strongly associated with a self-report of abuse or neglect of a child (Egami,
Ford, Greenfield, & Crum, 1996).
Although parental mental illness has not been causally linked to child welfare
involvement, it has been implicated as a predictor for entrance into the CWS. Kotch,
Browne, Dufort, Winsor, & Catellier (1999) found that maternal depression,
psychosomatic symptoms and the consumption of alcohol, among other characteristics
- assessed in infancy, all predicted entrance into the CWS in the child's first four years of
life. Similarly, Chaffin, Kelleher & Hollenberg (1996) utilized Wave I and Wave II data
from the ECA to assess risk factors for child maltreatment. Substance abuse disorders
were the most significant predictor of both physical child abuse and neglect, while
depression emerged as the most significant predictor of physical abuse, but not neglect.
Interestingly, depression was univariately related to neglect in this study; however, no
significant association remained once substance abuse was controlled for in the overall
hierarchical logistic regression model. The authors suggest that the relationship between
depression and· child neglect may be mediated by substance abuse, indicating that
psychiatric risk factors should not be considered in isolation.

Domestic Violence and Child Welfare System Involvement
A growing body of literature provides strong evidence that children who live in
households where domestic violence occurs are at risk for being maltreated. It is
estimated that approximately 6% of children in the U.S. are likely to be physically
maltreated in any given year in families in which marital violence occurs (Appel &
Holden, 1998). The rate of co-occurrence of domestic violence and child maltreatment
varies depending on the research methodologies utilized, as does the extent to which
domestic violence is linked with either physical child abuse, psychological child abuse,
neglect or sexual abuse. In a study that re-analyzed a sub-sample (N = 2,733) from the
1985 National Family Violence Survey, the presence of wife abuse, although significant,
explained less than 1% of the variance in the criterion variable of physical child abuse;
0.9% for physical punishment (not reaching the level of abuse) and 0.2% for verbal abuse
(Tajima, 2000). Research suggests that the link between domestic violence and all forms
of child maltreatment is stronger within families who are in the CWS for confirmed child
maltreatment. Rates of domestic violence among families in the CWS vary from 38% to

10

54% (Bowen, 2000; McGuigan & Pratt 2001 ).
Although domestic violence tends to be associated with all forms of child
maltreatment, the strongest association appears to be with physical child abuse. Rates
vary by study, yet generally, domestic violence more than triples the likelihood of
physical child abuse while the likelihood of all other forms of child maltreatment is
roughly doubled by the occurrence of domestic violence (McGuigan & Pratt, 2001;
Rumm, Cummings, Krauss, Bell & Rivera, 2000).

Parental Incarceration and Child Welfare System Involvement
More than 1.9 million people are currently incarcerated in prisons or jails- a
68% increase since 1990 (Beck & Kraberg, 2001). Nationally, incarcerated persons are
disproportionately people of color; 38% of prisoners are White (compared to 69.1% of
the general population); 45% of prisoners are African American (compared to 12.1 % of
the general population) and 17% of prisoners are Hispanic/Latino (compared to 12.5% of
the general population) (Beck & Harrison, 2001; Beck & Karberg, 2001; U.S. Census
Bureau, 2001 ). Forty six percent of incarcerated parents report living with their children
prior to incarceration and of the nation's 72 million minor children, 2.1 % had an
incarcerated parent in 1999. Estimates indicate that African American children are nearly
nine times more likely to have a parent in prison than are White children, and
Hispanic/Latino children are three times more likely to have an incarcerated parent, when
compared to White children (Mumola, 2000).
Data on the number of children who enter the CWS as a result of parental
incarceration are lacking, and there is considerable variability between states in their
efforts to gather data on the rates of parental incarceration for children in the CWS.
Estimates indicate that approximately 10% of the children of female prisoners and 2% of
the children of male prisoners are in some form of out-of-home care, yet it is unclear
whether these children were already in the CWS when their parent( s) were incarcerated
or how many entered the system specifically because their parent(s) were incarcerated
(Beck, Gilliard, Greenfeld, Harlow, Hester, Jankowski, Snell et al. 1992). A study
conducted by the U.S. Department of Health and Human Services (1997) concluded that
incarceration represented the main presenting problem in 4% of the cases in the CWS in
1994. In a survey ofall 50 states, the Child Welfare League of America (1998) reported
that only 5 of the 3 8 reporting states provided estimates of the percentage of children in
the CWS with incarcerated parents; these estimates ranged from 1.6% to 29 .5%.
Although the majority of prisoners are men, women prisoners are more likely to be
parents than are male prisoners and are also more likely to leave their dependent children
without a primary caretaker when they are incarcerated (Snell, 1994). Approximately
65% of incarcerated women have children under the age of 18, with an average of 2.11
children (Greenfeld, & Snell, 1999). The number of children entering the CWS due to an
incarcerated parent may be becoming a larger issue for child welfare agencies as the
number of women prisoners has risen 110% since 1990, so that in 2000, 156,200 women
were incarcerated (Beck & Karberg, 2001 ). Additionally, the impact of drug-related

11

activities on female prison population is also an area of concern for the CWS. Between
1990 and 1996, the number of female offenders serving time for drug-related offenses
doubled so that more than 1/3 of all female offenders in 1996 were serving time for drugrelated offenses (Gilliard & Beck, 1998).
Summary ofParent and Family-Related Factors and Child Maltreatment
While several studies point to a correlation between child maltreatment and the
parent and family-related risk factors of parental substance abuse, parental mental illness,
domestic violence and parental incarceration, there is a lack of empirical evidence to
support a causal connection. Additionally, the possible interactive effect of parent and
family-related risk factors on the disproportionate representation of children of color in
the CWS has not yet been studied. Unfortunately, most studies cited only used two or
three ethnic categories (e.g. White, African American and/or Hispanic/Latino), making it
difficult to draw conclusions on the role of ethnicity in parent and family-related risk
factors associated with CWS involvement. Despite these limitations, there is convincing
evidence to suggest that parent and family-related risk factors do play a role in CWS
involvement. Most importantly, findings suggest that parent and family-related risk
factors are likely related to one another in multiple and reciprocal ways. These risk
factors should not be considered in isolation, but rather as interacting factors that are
associated with CWS involvement and possibly-the disproportionate representation of
children of color in the CWS.
Santa Clara County Characteristics
In fiscal year 1997/1998 Santa Clara County served 4,775 clients over the age o/25
in their drug and alcohol programs; 44.2% of these clients were White, 39. 7% were
Hispanic/Latino, 10. 6% were African American, 4. 0% were Asian American/Pacific
Islander and 1. 4% were American Indian (State of California, Department of Alcohol
and Drug Program 1998, cited in Santa Clara County Public Health Department, 1999).
Santa Clara County has a rate of3.9 deaths per 100,000 due to drug use; deaths due to
drugs in Santa Clara County have declined steadily since the late l 980's and the county's
drug-related fatality rate is consistently lower than the overall rate for the State of
California (Santa Clara Valley Health and Hospital System, 1997). Additionally, in Santa
Clara County's Behavioral Risk Factor Survey conducted in 1997, 12.5% of respondents
were classified as binge drinkers and 6.2% as chronic alcohol abusers (Santa Clara Valley
Health and Hospital System, 1997).
Approximately 7% of the general population in Santa Clara County obtains
services from the mental health department (Santa Clara Valley Health and Hospital
System, 1997). Santa Clara County's Mental Health Department has a client base of
approximately 19,000 clients; 27% of these clients are under the age of 18; 64% are
between 18 and 59; and 9% are over the age of 60. Whites constitute approximately 49%
ofthe mental health system clients; African Americans 8.3%, Hispanic/Latinos 25.6%,
Asians 12.5%, Native Americans 1.2% and Other (non-whites) are 2.8% of the clients in
the mental health system. Thus, over 50% ofthe clients served by the county's mental

12

health system are persons ofcolor (Santa Clara Valley Health and Hospital System,
1997).

In the year 2000, the Santa Clara County District Attorney's Office reviewed 86
new reports of domestic violence each week-charges were brought in 3,076 of these
cases. In 24% of these cases, children were present during the domestic violence incident.
Since 1983, the number of Emergency Protective Restraining Orders (EPRO) have
increased seven fold in Santa Clara County-to a total of 1,866 EPRO in 2000. Fifty-one
percent of these EPRO's involved children in the home. In the year 2000, emergency

housing and shelter services for domestic violence victims served 4,217 clients; 41.6% of
these clients were Hispanic/Latino, 31 ..4% were White, 12.3% were Asian, 7.0% were
African American, 1.3% were Native American, 2.0% were "Other,, and 4.5% were of
unknown ethnicity. Eighteen people died of domestic violence in Santa Clara County in
2000 (Santa Clara Valley Health and Hospital System, 2001).

Santa Clara County houses the fifth largest jail system in California and the 14th
largest jail system in the United States. On an annual basis, the Santa Clara County
Department of Corrections books more that 65,000 persons who have been arrested and
there is an average daily population of 4,242 inmates in the County's six main
correctional facilities. These inmates have an average length of stay of approximately 93
days and 80% of the inmates are reported to have a history of drug and alcohol related
problems. Felony drugs charges are the most frequently reported (31 %) serious charge in
the Santa Clara County inmate population. Approximately 50% of the incarcerated
population has never served more than six months in custody while 35% have spent one
year or more in jail or prison. Current statistics on the ethnicities of inmates in Santa
Clara County facilities are unavailable (Santa Clara County Department of Corrections,
2001).
Bias in Initial Reporting and Subseguent Service Delivery
Research suggests that the disproportionate representation of children of color in
the CWS may be related to race and class bias in the initial reporting and subsequent
processing of children through the various phases of child welfare proceedings. This
possible differential treatment may be associated with many factors including bias on the
part of individual workers and/or ~tructural aspects of the CWS itself (Egami, Ford,
Greenfield & Crum, 1996). Most of the research in the area of bias and involvement of
children of color in the CWS has centered on issues related to the initial report.
Studies have indicated that less than 50% of reportable child maltreatment
situations are actually reported and that there may be racial and economic differences in
who reports, who gets reported and the types of maltreatment that are reported (Ards,
Chung & Myers, 1998; Ards & Harrell, 1993). Using data from the National Incidence
Study of Child Abuse and neglect (NIS 1-2-3) containing reported cases of child
maltreatment, as well as unreported cases obtained from a community sample, the NIS-3
findings corroborated results of the NIS-1 and NIS-2 (conducted in 1980 and 1986
respectively), in finding no race/ethnic differences in child maltreatment incidence

13

(Sedlack & Broadhurst, 1996). These findings parallel Ard's (1992) secondary analysis
of NIS-1 and NIS-2 data in which African American communities were found to have
lower rates of maltreatment than Caucasian communities once variables such as income
level, unemployment rates and the extent to which an area is urban or rural were
statistically controlled.
Secondary analyses of NIS-1 and NIS-2 data on the types of agencies reporting
child maltreatment revealed that, in 1986 CPS was more likely to be aware of hospital
cases involving African American children, than Caucasian children, controlling for type
of abuse (Ards, & Harrel, 1993). Additionally, Hampton and Newberger's (1985)
secondary analysis of NIS-1 data collected in 1980 indicated that compared to other
agencies, hospitals were more likely to report low SES African American and
Hispanic/Latino families than Caucasian families.
Research on the impact of case characteristics on child abuse reporting decisions
has shown that race and SES are variables that are inconsistent, although at times
significant, factors in report decision-making (Zellman, 1992). In one study, physical and
sexual abuse vignettes depicting lower SES and African American families were
typically determined to be more serious, more likely to be defined as abuse, and more
likely to be perceived as requiring a report under the law. Outcomes were generally
judged to be better for lower status African American families, and although not
consistently reaching significance across all vignette categories, respondents were more
likely to report African Americans. Interestingly, when race was varied in the neglect
category, the benefit of the report was perceived as higher if the child was Caucasian and
of a higher SES, rather than African American (Zelman, 1992).
It is also possible that child welfare agencies substantiate reports at a higher rate
depending on the reporter, the perpetrator, or family-related characteristics. While there is
a lack of research on reporting characteristics and substantiation of child welfare cases,
one study has suggested that substantiation rates for neglect are higher in impoverished
communities (Drake & Pandy, 1996). More research on cases that are reported and
substantiated and characteristics that distinguish them is necessary in order to understand
ways in which bias may play a role in the initial involvement of children of color in the

cws.

Research suggests that once in the system, families of color receive fewer services
and have poorer outcomes than their White counterparts (see Courtney, Barth, Duerr
Berrick, Brooks, Nedell & Park, 1996 for a review). A six-year longitudinal study of
children in California found that African American children were far less likely than
Caucasian or Hispanic/Latino children to be reunited with their families, and less likely
than Caucasian children to be adopted. Hispanic/Latino children were also found to be
more likely than Caucasian children to remain in care than to be adopted (Barth, 1997).
While research in the area of bias in initial reporting and subsequent service delivery is
scant and inconclusive, differences in service provision may reflect biases in the ways in
which children and families of color are assessed and subsequently provided services.
Further research on bias and the role it plays in the involvement of children of color in

14

the CWS is necessary.
Santa Clara County Child Welfare System haracteristics
Focus group results from Phase 1 of this study indicate that workers from Santa
Clara County's CWS do perceive differential treatment at various phases in the CWS.
When asked what they felt were the factors related to the disproportionate representation
of children of color in the CWS, focus group participants noted several reasons,
including: over and under reporting by mandated reporters; bias on the part of ER
workers regarding which children should be removed; lack of agency support for
Informal Family Service and Voluntary Family Maintenance services; DI workers being
more likely to detain children of color than White children; and African American and
Hispanic/Latino cases being by-passed for services more frequently than other groups.
Phase 1 analysis of CWS/CMS data also corroborated the notion that children of
diffe~ing ethnicities receive differential treatment in the various stages of the CWS.
African American children in Santa Clara County's CWS spent significantly more time in
placement than their White, Hispanic/Latino, Asian American/Pacific Islander,
Vietnamese, and Filipino peers. Also, following placement in a relative home, the second
most frequent placement for African American, Native American, White and
Hispanic/Latino youth, was a Foster Family Agency (FFA) while the second most
frequent placement for Asian American/Pacific Islander youth was a Foster Family Home
(FFH). (Please see Phase I Final Report for a complete description of findings.)
Poverty and Characteristics of Impoverished Communities that
Increase the Risk of Children of Color Entering and Staying in the Child Welfare System
While initial conceptions of child maltreatment focused on medical and
psychiatric factors and tended to downplay social and socioeconomic factors (Kempe,
Silverman, Steele, Droegemueller & Silver, 1962; Steele & Pollock, 1968), there is
currently considerable evidence that cases of child maltreatment have been
disproportionately found among low-income and poor families (Colton, Corbin, Su and
Chow, 1995; Drake & Pandy, 1996; Garbarino & Sherman, 1980; Gelles, 1992; Pelton,
1978; Zuravin, 1989).
This body of evidence is particularly relevant to our research question because
African American and Hispanic/Latino communities tend to have higher rates of poverty
than other racial/ethnic groups. U.S. Census figures indicate that approximately 19
percent of all American children under age 18 live in families who are below the official
poverty line. Of this 19 percent, approximately 26 percent are Hispanic/Latino and
approximately 26 percent are African American children. Thus, over 50 percent of all
American children living in poverty are Hispanic/Latino or African American (U.S.
Census Bureau, 1999). African Americans also tend to spend longer amounts of time
living in poverty and have the lowest exit rate from poverty and the longest median
poverty spell of any other racial/ethnic group (Duncan, 1991; Naifeh, 1998).
Additionally, Hispanics/Latinos have the highest episodic poverty rate of any other ethnic

15

group and Hispanic/Latino families with children under 18 are twice as likely to live in
poverty as non-Latinos (Naifeh, 1998; Zambrana & Dorrington, 1998).
The links between poverty and outcomes related to child well-being, including
child maltreatment rates and entrance into the CWS, are difficult to isolate because a
number of family and neighborhood conditions often occur simultaneously with poverty.
The influence of poverty has been shown to operate both through restricted resources to
individual families and through macrostructural forces that shape impoverished
communities (Coulton & Pandey, 1992). Poor families are more likely to be headed by
young females with low levels of educational attainment that are unemployed or in the
low-wage market (Brooks-Gunn, Duncan, & Maritato, 1997). Moreover, neighborhoods
experiencing concentrated poverty (e.g. neighborhoods where the poverty rate exceeds 40
percent) often possess accompanying negative characteristics such as high crime rates,
neighborhood violence, poor public schools and dilapidated housing conditions (BrooksGunn et al., 1997).
Research into neighborhood and community factors related to child maltreatment
is currently fairly underdeveloped (Drake & Pandey, 1996). While some studies have
demonstrated that concentrated poverty is related to higher rates of child maltreatment
(Garbarino & Sherman, 1980; Steinberg, Catalano & Dooley, 1981) others have found
that economic deprivation is not the sole factor producing negative outcomes for children
(Ards, 1992; Coulton & Pandey, 1992). Other factors found to pose extreme risk to
children and adolescents are high concentrations of female-headed households, high
crime rates, and high concentrations of families living in public housing. These
conditions may affect parenting behaviors that, in turn, produce poorer health and
developmental outcomes for children in concentrated poverty area (Ards, 1992; Coulton
& Pandy, 1992).
In an effort to explain the mechanisms through which concentrated poverty may
affect child maltreatment rates, Coulton, Korbin, Su & Chow (1995) and Coulton, Korbin
& Su (1999) investigated the mediating role of a community's level of social
organization. The authors describe community social organization as including factors
such as a community's economic status, residential mobility, family structure and
ethnicity. Using census and child welfare agency data for 177 urban census tracts,
Coulton et al. (1995) found child maltreatment rates to be related to a number of
structural determinants of community social organization including poverty, excessive
numbers of children per adult resident, household and age structure, geographic
proximity of neighborhoods to concentrated poverty, population turnover and a
concentration of female-headed households. Race/ethnicity was not examined in this
study.
Community social organization may also be related to neighborhood racial
segregation-or the over/under representation or concentration of certain ethnicities in
specific geographic locations. Racial segregation of neighborhoods, like the variable of
poverty, has been linked to poor outcomes for communities of color (Massey, 1993).
Nationally, it is estimated that one third of African Americans live under conditions of

16

intense racial segregation. For Hispanics/Latinos, low to moderate levels of racial
segregation are found in Hispanic/Latino communities across the nation (Massey &
Denton, 1989). In a review of Census tract data, Enchautegui (1997) found that
Hispanics/Latinos living in neighborhoods that were comprised of predominately other
Latinos had poverty rates over seven percent higher than Latinos in the national
population and had significantly higher unemployment rates than Latinos in the general
population. Hispanics/Latinos living in racially segregated communities were also I I%
more likely to live in female-headed households than Latinos in the general population
and four percent more likely to not have a high school diploma. These data suggest that
the negative effects of racial segregation do impact both African American and
Hispanic/Latino communities.
Research supports the association between concentrated poverty, neighborhood
context, the mediating role of a community's level of social organization and the
occurrence of child maltreatment and subsequent contact with the CWS. The increased
likelihood of experiencing poverty, and accompanying low levels of community social
organization, among communities of color may be a contributing factor to the
disproportionate involvement of children of color in the CWS. The overrepresentation of
African American and Hispanic/Latino children living in economically deprived
households not only increases their risk of coming into contact with the CWS, but also
increases risk for poor outcomes on assessments of health, cognitive development, school
achievement and emotional well-being, as well as for peer conflict, depression and low
self-confidence (Brooks-Gunn, Duncan & Maritato, 1997; Huston, 199 I). Additionally,
in an investigation of the cumulative psychosocial stress of living in poverty, McLoyd
and Wilson ( 1991) found that mothers' mental health worsened as a function of her
worsening economic situation. As this process occurred, mothers in the sample became
less likely to behave supportively and positively towards their children and were less
satisfied with their parenting role. The resulting outcome for children of these mothers
was higher levels of anxiety and depression.
Current research on these associations has focused almost exclusively on African
American and Hispanic/Latino communities, both of which are at an increased risk of
living in poverty. Yet although these communities may share an increased likelihood of
experiencing poverty and the detrimental effects of poor community social organization,
the mechanisms through which African American and Hispanic/Latino families become
impoverished and the characteristics of their respective communities may differentially
impact their entrance into and experiences with the CWS.
Santa Clara County Characteristics
Slightly more than 8% ofpeople of all ages in Santa Clara County are living in
poverty and approximately 13.4% of children are living below the poverty level (U.S.
Census, 2001 ). Additionally, in I 998-1999, 31 % of children in Santa Clara County were

eligible for subsidized school lunches (Kids in Common, 2000). Households that are
already in the lowest 20 th percentile of Silicon Valley's income distribution have not seen
their average income earned (adjusted for inflation) increase at all since l 993~espite

17

the fact that during the same time period the cost of living has increased 20%. At the
same time, incomes for the top-earning 20% ofhouseholds rose an estimated 20%
(adjusted/or inflation) since 1993-indicating that the gap between the wealthy and the
non-w_ea/thy is extremely wide and growing in the Silicon Valley area (Joint Venture
Silicon Valley, 2001).
The difficulties experienced by households in the bottom of Silicon Valley's
income distribution, are also reflected in the fact that, in 2001 only 15% of Santa Clara
County houses that were sold were affordable for households with the median income
(which is approximately, $64,000)-a figure that is significantly lower than the national
average of 63% (Joint Venture Silicon Valley, 2002). Not surprisingly, homelessness is a
problem in Santa Clara County. In the Santa Clara County Homelessness Survey
(utilizing a non-random self-reporting sample) conducted in 1999, it was estimated that
there are 20,000 homeless persons in Santa Clara County-one-third ofwhom were
children (Burstein & Woodsmall, 1999). Hispanic/Latinos represented 36.2% of those
surveyed, 34.9% were White, 17.3% were African American, 4.3% Native American,
3.3% Asian American/Pacific Islander and 4.0% were "Other" or refused to answer.
Thirty-five percent of those surveyed were employed; and 25.6% reported being
homeless for 1 to 3 months, 22.7% for 6 to 12 months; 20.l % for I to 2 years and 17.8%
had been homeless for 2 to 5 years (Burstein & Woodsmall, 1999).
In 2001, the Silicon Valley area lost 25,000 jobs and average annual pay in
Silicon Valley dropped by an estimated 2% (Silicon Valley Joint Venture, 2002). Yet
even for parents who are employed, it can be difficult to survive in Santa Clara County
due to the high cost of living. For instance, 50% of the fastest growing jobs in the County
pay less than $10 an hour (Santa Clara County Children and Families First Commission,
2000). Yet estimates have shown that a family with two working adults and two children
must earn a minimum of $26.20 an hour to meet minimum living costs in Santa Clara
County (Kids in Common, 2000). These figures suggest that numerous children and
families in Santa Clara County likely experience economic hardships.
The large immigrant population in Santa Clara County may be the hardest hit by
the recent economic downturn. According to recently released figures from the U.S.
Census Bureau, the immigrant population in Santa Clara County grew at a faster pace
than the rest of California between 1990 and 2000. Currently, 34.1% of Santa Clara
County's population is foreign born compared to 11% nationally and 26% at the state
level (U.S. Bureau of the Census, 2002). For immigrants who come from poor countries
with scant educational opportunities, job prospects can be scarce in a region
characterized by a highly technological culture.
Levels of community social organization are difficult to measure and systematic
efforts to describe determinants of community social organization in Santa Clara County
are lacking. In one recent analysis of census tracts within the City ofSan Jose, a distinct
pattern ofseparation between Whites and people of color was noted (City ofSan Jose,
2001). The number of census tracts in which 70% to 100% ofthe popu(ation is comprised
of one ethnic group are as follows: 8 census tracts are predominantly Hispanic/Latino; 2

18

are Asian American/Pacific Islander; and 21 are White (City of San Jose, 2001).
Although racial/ethnic segregation exists in the Santa Clara Valley, the ways in which
poverty, community social organization and racial segregation affect CWS populations in
this locality are largely unknown.
The Impact of Recent CW Policy Initiatives on the Involvement of
Children of Color in the Child Welfare System
Since passage of the Indian Child Welfare Act (ICWA) in 1978, only two Child
welfare policies have been specifically targeted to children of color in the system. The
Multi-Ethnic Placement Act (MEPA) of 1994 later amended by the Interethnic Adoption
Provisions (IEP) in 1996 was intended to accomplish the following: decrease the time
children wait to be adopted; promote recruitment and retention of diverse foster and
adoptive parents; and eliminate discrimination based on race, color, or national origin in
placement decisions. The enactment ofMEPA-IEP has several potential implications for
children of color. Research indicates that African American children are severely overrepresented in the population of children waiting for a permanent home (Hollinger,
1998). IfMEPA-IEP is implemented as was intended, children would spend less time in
the system waiting for permanent placements, thereby preventing the adverse effects of
long-term institutional care. Although MEPA-IEP prohibits agencies to consider culture
or ethnicity in placement decisions, the act requires that States and agencies receiving
Federal funds actively recruit foster and adoptive parents to reflect the cultural and ethnic
backgrounds of the children in their agencies. More foster and adoptive families of color
available to take in children of color without delaying the permanency placement process
would potentially be of benefit to children of color waiting to be adopted (Knapp,
McDonald & Diamond, 2001). MEPA-IEP also prohibits the removal of a child from a
safe and stable trans-racial placement in order to place a child with a family of similar
ethnic background. This would be of benefit to children of color as fewer disruptions in a
child's life have continually been found to contribute to more positive outcomes.
To date there has been no research on the impact ofMEPA-IEP on children of color in
the CWS. It remains unclear whether this act has been of benefit to children of color or
whether several inherent obstacles to the implementation ofMEPA-IEP, such as
confusion about the law, lack of resources for recruitment of foster and adoptive parents,
resistance from workers, and fear of litigation, have resulted in the inappropriate and
unsuccessful implementation ofMEPA-IEP.
The Adoption and Safe Families Act 1997(AFSA) represented a key shift in public
policy by emphasizing the primacy of protection of children over preservation and
reunification of the family and creating provisions that were intended to expedite
permanent placements for children in out-of-home care by making it easier to remove
children from dangerous home environments (U.S. Department of Health and Human
Services, 1999). At the same time, the act included provisions for continued funding of
family preservation services as well as kinship support services - programs designed to
strengthen families so that children can either stay in the home or be reunified without
delay. To date, there have been no systematic studies on the impact of AFSA and its

19

provisions on children of color. ASFA's primary provisions include shortened timelines
and support for family preservation and kinship services, bypass criteria, concurrent
planning and adoption incentives.
ASF A mandated new shortened time lines for the provision of reunification
services for children in out-of-home care, as well as for the termination of parental rights
(TPR) process. The parental risk factors and characteristics of communities of color that
may increase chances of entering the CWS, explicated in the first section of this paper,
will also likely impact the effectiveness of reunification and other support services once
families become involved in the CWS. These circumstances in combination with a lack
of necessary social services for reunification, the enormous workloads of individual
social workers and shorted time frames may create overwhelming barriers to successful
reunification for children of color in the CWS. Families trying to meet reunification
requirements need a wide array of social services to address these myriad concerns.
Under ASF A, reunification and related support services need to address many of these
co-existing problems within six and 12-month time limits. Yet, data suggest that
frequently needed services are simply not available or are inadequate in meeting the
multifaceted needs of families in the CWS (U.S. Department ofHealth and Human
Services, 1999; U.S. General Accounting Office, 1998). Without comprehensive social
services, the shortened time lines for reunification may be not sufficient for parents to
meet all of the conditions for reunification.
In addition to shortened time frames, ASF A introduced new "bypass criteria," or
circumstances under which a state is relieved of the requirement to make "reasonable
efforts" for reunification. Bypass criteria focus on areas such as parent substance abuse,
parent criminal history and parental mental disability and are intended to ensure child
safety and to expedite permanency planning efforts by quickly identifying families for
whom reunification is unlikely (U.S. Department of Health and Human Services, 1999).
Ostensibly, the early identification of such families will free up more children for
adoption or alternative permanent placement and shorten their length of time in the CWS.
Once the court verifies the presence of a bypass criterion, the state must begin within 30
days to find the child an alternative permanent home (U.S. General Accounting Office,
1999).
While there is no extant research on the impact of bypass criteria on children of
color in the CWS, given certain characteristics of communities of color that increase the
risk of children of color entering and staying in the CWS, the use of bypass criteria may
have numerous potential implications for children of color in the CWS. Parent-related
correlates of maltreatment described in the first section of this review, including
substance abuse, mental illness, domestic violence and incarceration together with the
macro-level correlates of child maltreatment such as chronic poverty; lack of community
social organization or racial segregation may all increase the likelihood that a child of
color entering the CWS will possess a bypass criterion. Additionally, the lack of an
empirically validated method for assessing the presence of a bypass criterion could result
in more children of color being inaccurately assessed as possessing a bypass criterion.
More extensive empirical evidence on the impact of bypass criteria on families and

20

children of color is sorely needed.
ASFA's goal of timely permanence is also reflected in the practice of concurrent
planning-a service strategy that calls for the creation of two service plans for every
child entering the CWS: a reunification plan and an alternative permanency placement
plan. Concurrent planning practices could potentially have numerous effects on children
of color in the CWS. Concurrent planning creates dual roles for social workers who are
supposed to simultaneously reunify families while also seeking an alternative permanent
placement. This dual role may make it more challenging for social workers to provide
reasonable efforts to reunify. As such, alternative permanent placements may be given
priority over reunification. While there has been no research on the impact of concurrent
planning on. children of color in the CWS, these factors have raised concern that children
of color will be "removed even more precipitously from their families and communities
for permanent placement elsewhere" (Katz, 1998, p.6).
Finally, ASFA created a new category within Title IV-B entitled the "Adoption
Promotion and Support Services Program." This new program provides states with cash
incentives for each child in foster care who is placed in a permanent home. Under ASF A,
a state receives $4,000 in federal funds for each foster child adoption that exceeds the
previous year's foster child adoption rate·. States also receive an additional $2,000 for
each special needs adoption. The Adoption Promotion and Support Services Program
stipulates that states are to use these federal funds to provide post-adoption services to
children and families (U.S. Department of Health and Human Services, 1999).
The Personal Responsibility and Work Opportunity Reconciliation Act
(PRWORA) of 1996 fundamentally changed America's welfare system, replacing the Aid

to Families with Dependent Children program (AFDC) with the Temporary Assistance to
Needy Families (TANF) program. This change ended low-income families' entitlement
to receive cash assistance from the federal government by creating a block-grant program
to the states. Under TANF, most recipients are required to work while receiving benefits
and are limited in the amount of time they may receive assistance. The legislation also
makes persons convicted of a· drug-related felony permanently ineligible for both T ANF
and food stamp assistance, requires minor parents to live at home to receive assistance
(unless the state agency determines that the minor parent has been subjected to abuse or
exploitation), and makes immigrants arriving after the passage of PRWORA ineligible
for federal means-tested benefits for a period of five years.
Nationally, more than half of children in foster care come from homes that are
eligible for welfare - a figure that represents an increase in welfare-eligible families in the
CWS from 11 percent in 1970 to 53 percent in 1996 (U.S. House of Representatives,
1998). Although PRWORA made few changes to federal child protection programs
specifically, a number of the changes it instituted are likely to affect states' child welfare
systems due to an overlap in the child welfare and welfare populations, as well as
changes in funding streams used by child welfare agencies. While proponents of welfare
reform suggest that the new requirements will help low-income families achieve selfsufficiency more quickly, it is not clear what will happen to families who cannot meet the

21

new requirements and will either lose benefits or have them reduced. As mentioned
earlier, the likelihood of being referred to the CWS is correlated with low-income status
and factors related to poverty. Consequently, if the economic well being of families
decreases, reports of abuse and neglect may increase (U.S. Department of Health and
Human Services, 1998).
It remains unclear how welfare reform has affected the CWS and in particular,
children and families of color involved in the system. Preliminary findings on the impact .
of welfare reform on American families indicate that between 1997 and 1999, poverty
rates declined, food hardship for low-income families decreased, and the rate of twoparent families increased (Moore & Vandivere, 2000; Zedlewski, 2000). These national
gains, however, do not hold true for either African American or Hispanic/Latino families.
National estimates indicate that while employment rates of African American parents
rose, African American families experienced no decrease in rates of poverty, food
hardship or the incidence of single-parent families and their housing hardship worsened.
Between 1997 and 1999, Hispanic/Latino families saw decreases in poverty and the rate
of single-parent families, but experienced declines in health status and health insurance
(Staveteig & Wigton, 2001; Weil & Feingold, 2002). These increasing racial and ethnic
disparities could imply that welfare reform and related public policies are working better
for whites than for children and families of color. It is possible that the disparities seen as
a result of welfare reform will carry over and affect children and families of color in the
CWS who, as we have described earlier, may already be at a disadvantage due to the new
ASFA regulations. However, further research on the impact of welfare reform on
American families is required before definitive conclusions in this area can be reached.
Santa Clara County Characteristics

The impact of recent child welfare policies on children of color in Santa Clara
County has not yet been systematically investigated. As noted above, there are numerous
possible implications of these policies on children of color in the CWS. Phase 1 results
indicated that focus group participants did perceive that some aspects of child welfare
policies are related to the disproportionate representation of children of color in the CWS.
Among other factors, focus group participants felt that children of color in the CWS were
negatively impacted by bypass criteria, strict timelines imposed on families by the courts
and the need for more training for workers about AFSA.
The impact of welfare reform on children and families in Santa Clara County is
difficult to ascertain. It is notable that ofthe County's 27,415 Ca/Works participants,

44.2% are Hispanic/Latino, 34.8% are Asian American/Pacific Islander (82.5% ofwhom
are Vietnamese), 10. 6% are White, 6.4% are African American and 0. 6% are Native
American (Lightbourne, 2001). In an effort to better serve clients who are both in the

CWS and receiving some form of welfare, the Department of Family and Children's
Services (DFCS) and the Department of Employment and Benefit Services (DEBS) have
collaborated to place Employment Technicians (ET) in the DFCS unit. ET's work with
child welfare families to identify eligible participants and work to integrate the DFCS
service plans with the participants' self-sufficiency plans (Santa Clara County, DEBS,

22

2001). The outcomes of these efforts and the impact of welfare reform in general on the
disproportionate rates of children of color in the CWS are unknown at this point in time.
Summary
This research review suggests that factors related to the disproportionate
representation of children of color in the CWS are multiple and complex. Although
definitive findings from the research literature are lacking, it is clear from our review that
parent and family factors, race and class biases, social factors, and public policies all
impact CWS involvement--yet these factors should not be viewed in isolation from one

another. Research repeatedly suggests that these risk factors are often inter-related in
ways that are not fully understand or researched at this point in time. Additionally, the

factors related to the under-representation of certain ethnicities, including Asian
American/Pacific Islanders and Whites have virtually been ignored by the research
literature. We do not know if there are certain characteristics of these ethnic groups that
serve to protect them from child maltreatment-or perhaps that there is an under
reporting of child maltreatment in these communities.
Data related to the characteristics of children, families and communities in Santa
Clara County suggest that this locality does possesses risk factors for CWS
involvement-and that these factors are also inter-related. Most notably in Santa Clara

County, Hispanic/Latinos are over-represented in drug and alcohol rehabilitation
programs, as well as in domestic violence shelters and the homeless population
(compared to their population in the county); over 50% ofmental health clients in Santa
Clara County are people of color; and 80% ofSanta Clara County's inmate population
have a history ofdrug and alcohol problems. Additionally, there is an ever-widening gap
between the wealthy and the non-wealthy in Santa Clara Valley-and the fastest growing
occupations in this area pay less than $10 an hour-a figure far below current estimates
ofthe minimum living costs in Santa Clara County. Lastly, findings from Phase 1 suggest

that children of color do receive differential treatment at various phases on the CWS and
that recent public policies are also related to the disproportionate representation of
children of color in the CWS. Yet, the particular ways in which risk factors for entering
and staying in the CWS operate in Santa Clara County require further explorationssome of which are reported in sections V-VIII of this report. The next section provides an
overview of the specific methods and procedures used in Phase 2.

23

IV. Phase 2 Study Methods and Procedures
The primary methodology for Phase 2 of our study involved extensive, in-depth
reviews of 403 closed child welfare case records, a parallel descriptive analysis of 1720
closed cases within the CWS/CMS database, and key informant interviews with
managers and supervisors in the county's Department of Family and Children Services
(DFCS).
Case Record Review - Procedures
From 6761 total case closures over an 18-month period, a data file containing
1753 cases representing one child per family and one case opening was constructed. This
data file was used to obtain our target sample of 400 cases selected randomly, guided by
stratification according to ethnicity, age and service type. See Attachment 1 for a
description of the study sampling procedures.
The research team also worked closely with DFCS on accessing case files and
setting up the logistics for the case record review data collection process. See
Attachment 2 for a description of case record review data collection process and
procedures.
In order to record and code data from the case files, it was necessary to develop
an extraction form that assessed the variables of interest. During this study period, the
CWRT developed the case record extraction form to be used for data collection (see
Attachment 3). Beginning November 26, 2001 a pilot test was conducted with each of
the 10 research assistants coding the same 4 cases, using the newly developed case record
extraction form. In order to ascertain the consistency of coding among the case reviewers,
a reliability study was conducted on the pilot cases. See Attachment 4 for a description
ofprocedures and results ofthe reliability study. Because some of the variables of
interest came from sources other than the case record review (including CWS/CMS,
personnel files, and eligibility data), we developed a table outlining the key variables and
their source. See Attachment 5 for the Variable Checklist.
In order to maximize study validity, we elected to use the court report section of
the case records for our case record·data extraction. The CWRT met with Judy Bushey of
DFCS to discuss the match between the content of the court reports and our data
collection demands. Ms. Bushey also provided invaluable assistance with the logistics of
the data collection procedures. The research assistants were trained by Stan Lee on the
organization and content of the court reports in two separate trainings held on October
29th and November 5th, respectively. In addition, Dr. Lonnie Snowden from the School
of Social Weifare at UC Berkeley, provided training on research methods and procedures
for the rese~rch assistants.
Data collection for the case record review portion of the study began on Dec. 17,
200 I and was completed on May 31, 2002. In order to ascertain the consistency of

24

coding among the case reviewers an initial reliability study was conducted on the cases in
January 2002. Results indicated that for the cases considered in the reliability study, 64%
of case record review items were found to have a good level of inter-rater reliability.
Thus, 80% of ten reviewers recorded with perfect agreement on almost two-thirds of
items assessed included on the case record review protocol.
During the data collection period, we continued to assess the reliability of the case
record review. We selected a subsample of forty reviewed cases in a manner such that
reviewers were unaware of which cases had been selected. We then had the assistants
perform a "blind" second case record review on those cases, without consulting the
original completed data collection form. Using methods described in attachment 4 we
then assessed levels of agreement between the first and second reviews. This process will
enable us to ascertain levels of reliability of recorded information at the heart of our
study. Analysis on data pertaining to the reliability study is in its final stages and will be
included as an addendum to this report.

Case Availability Status Analysis
Although the sampling frame for the case record review consisted of 1753 cases,
1298 cases had to be requested from Santa Clara County's Social Services Agency in
order to obtain 403 codable cases for the case record review. Of these 1298 cases, 529
(40.8%) were missing and/or unavailable for coding, 366 (28.2%) were identified as
voluntary family maintenance, and 403 (31.0%) available and coded for the case record
review component. Some files originally sampled but missing were located during our
data collection phase and thus included in the 403. VFM cases were excluded from the
case record review due to our interest in following cases through court proceedings and
closure. A comparison of missing versus VFM versus coded cases is included in
Attachment 6 of this report.
Case Record Review - Data Analysis

Close-Coded Data
A total of 403 cases were reviewed in-depth. The information recorded by 10
CWRT student research assistants on the data extraction forms were also entered by them
into the computer using SPSS, a statistics computer program. Primarily closed-coded
information (those variables with specific choices, e.g., gender, type of abuse/neglect,
grade level, marital status, etc.) was entered at this stage. This information was compiled
for analysis purposes, and is reported in Section VII of this report. This section contains a
descriptive analysis of the 403 cases and a comparison of key individual, family and
system-related characteristics across ethnic groups.

Open-Coded Data
Three variables were coded from open-ended questions on the data extraction
form: ( 1) criminal history of child, mother, father, alternate caretaker and perpetrator, (2)

25

services ordered for the child, mother, father and alternate caretaker at the initial
jurisdictional/dispositional hearing, and (3) case changes that were noted in the court
records.
A combination of qualitative and quantitative methods was used to code and
analyze these variables. Specifically, a standardized grid was created for each of the three
variables in which key characteristics were categorized. Criminal history was coded using
a standardized grid in which the most commonly occurring types of crimes were noted, as
well as the number and level of seriousness of the crime (e.g. citation, misdemeanor,
felony, etc.). The average number of criminal episodes was then generated from these
data. Similarly, the grid for ordered services listed the most commonly occurring ordered
services for each individual. The frequency of services was then generated from these
data.
A grid that listed categories of possible case changes was also created in order to
capture the types of changes over the case history. These categories were created on the
basis of their frequent occurrence in the court records. Initial categories included changes
related to placement, case status, services, visitation, criminal activity, health, personal
adjustment, mental illness, substance abuse, domestic violence, education, family and
parental issues, and financial situation.
For statistical analysis purposes, these categories were then collapsed into 3
broad categories. The first category was Placement and Placement Related Changes,
which included specific changes in status and placement (e.g., FM, FR, Bypass, PP),
custody, runaways, emancipation, and dismissals. The second was Service Related
Changes, which included visitation rights, orders for counseling, family services,
treatment, and other referrals, and compliance with service plan. The third category was
Psychosocial Status Related changes, which included changes in health, mental health,
disability, education, economic, criminal, substance abuse, family violence, cultural, and
other adjustment conditions. Within each of these 3 categories, changes were then coded
positive, negative, or other. Others are changes which could not be determined as either
positive or negative, for example "mother got married to man not the father" could
possibly be positive if this created a more stable home environment, or possibly negative
if this created more disruption in the home given a tenuous relationship between the child
and the new father. The frequencies of these changes were then generated.
This open-coded information was compiled and added to the analysis of the 403
cases, and is included in Section VII describing this sample, comparing key
characteristics across ethnic groups, and identifying factors that delineate these groups.
CWS/CMS Database of Closed Cases - Procedures and Data Analysis
Judi Boring and her staff were extremely helpful in providing data and technical
support regarding the collection and analysis of the larger 1753 case sample mentioned
above. A substantial amount of data was available through CWS/CMS and valid
information was available for I 720 cases. For example, specific allegations, demographic

26

descriptors, hearing dates, placement changes, and placement types were generally
accessible. However, due to the large volume of data, a ~omplicated series of computer
procedures had to be done (and are still being conducted) in order to synthesize key
information into analyzable form.
A parallel, descriptive analysis of this key information was conducted to help put
the case record reviews into perspective, and is included in Section VI. This section is
both a descriptive analysis as well as a general comparison using bivariate statistical tests
of case characteristics across ethnic groups.
Key Informant Interviews - Procedures and Analysis
(Please see section VIII of this report for a more detailed description of the methods and
procedures used in conducting the key informant interview portion of this study)
The key informant interview portion of this study was designed with input from
the project advisory board. The aim of this qualitative component was to explore the
perceived dynamics, factors, policies, and possibl~ solutions associated with the overrepresentation of children of color in the child welfare system. Eight in-depth interviews,
of approximately one hour in length, were conducted with key informants in managerial
positions in DFCS.
Key informants were selected based on their overall experience in child welfare
and capacity to address the representation of children of color at specific choice points in
different parts of the child welfare system. Two interviewees from South County, where a
vertical case management model is employed, provided qualitative data across different
facets of the child welfare system.
Qualitative methods of data analysis were used to identify themes that were
common across interviews. Please see Section VIII for a detailed description of the key
informant interview findings.
Note: The results presented in sections V-VII of this report focus on ethnic differences
across four groups: African American, White, Hispanic/Latino and Asian
American/Pacific Islander. In some analyses, when sample size permitted, an "other"
category was also included. The category designated as "other" consisted of Native
Americans, Ethiopians, White-Armenians, and White-Middle Easterners.
Special Note on Native Americans and Other Groups: Of the I 720 cases valid for the
general sample, 15 were Native American. Among the 403 cases reviewed in-depth, 4
were Native American. Given these relatively small numbers and our current
methodology focusing on identifying and generalizing key characteristics, Phase 2 's plan
was unable to include analyses that distinguished this ethnic group. This situation was
similar for many ethnic and special interest groups that had to be combined within major
categories, (e.g., the use of 5 major ethnic categories of African American, White-

27

European American, Hispanic/Latino, Asian American/Pacific Islander, and Other). The
CWRT acknowledges that all ethnic groups are important and that identifying each
group's experience is important. However, given the methodology for Phase 2 of this
study, it was not possible to conduct detailed analyses of smaller subgroups.

28

V.

Pathways Through the Child Welfare System: Case
Studies by Ethnic Group

The following section provides narrative descriptions of ways in which pathways
through the CWS differ by ethnic group. Guided by findings from our quantitative
analyses of data extracted from 403 case records (results presented in Sections VI and VII
of this report), we selected two cases that we felt represented typical pathways for each of
the four racial/ethnic groups (i.e., African American, White, Hispanic/Latino, and Asian
American/Pacific Islander) focused on in this report. Within each racial/ethnic group, one
case study illustrates the pathway for a case assigned to Family Reunification.at the
Jurisdictional/Dispositional hearing and the other the pathway for a Family Maintenance
case.
The eight case studies presented in this section reflect pathways and accompanying
individual, family and system-related characteristics that are substantiated by results
presented in sections VI and VII of this report. Although ethnicity may be one factor
associated with differences in pathways through the system, it is important to note that
there may be numerous other factors associated with the outcomes of the case examples
provided. The following case studies are presented solely as a means of illustrating our
findings, and giving the reader a preview of the results to follow.
Note: all names are fictional and any identifying information has been changed to
protect confidentiality
African American
Family Reunification Case

Samuel was a 5-year-old African American male who entered the CWS as a result
ofgeneral neglect, emotional abuse and caretaker absence/incapacity on the part ofhis
mother. This maltreatment was reported to the CWS by a hospital social worker. Samuel

had six prior referrals to the CWS and was receiving informal supervision services when
this maltreatment was reported. He had four siblings-two of whom were in the CWS.
Samuel was born in San Jose, California and was in kindergarten at the time of
the incident. He had been living with his mother and two siblings.

Nancy, Samuel's mother, was a 25-year-old African American female. She was
born in Monterey, California, and was divorced from her husband-Samuel's father.
Nancy was unemployed at the time of the incident; she had an 11th grade education, and
vocational training to be a nurse's aid; she was also receiving AFDC. Nancy had a history
of substance abuse problems, including both alcohol and cocaine. Nancy's criminal
history included one arrest ·for perjury, two convictions for fraud to obtain aid, one
conviction for assault and one conviction for prostitution.

29

John, Samuel's father, was a 32-year-old African American male. John wa,s
incarcerated in a state prison at the time Samuel entered the CWS. John had a criminal
history that included arrests for possession of a narcotic controlled substance, being under
the influence of a controlled substance -in public, battery, vandalism, and a vehicle code
violation. John also had two convictions for assault and battery, one conviction for
obstructing and resisting a police officer, and one conviction for driving under the
influence.

At the Jurisdictional/Dispositional hearing, Samuel's case was assigned to family
reunification services and he was placed with his maternal grandmother. Nancy was
allowed weekly, one-hour, supervised visits with her son, and she was also ordered to
attend a parenting class. John was not ordered to receive services.
After 12 months, family reunification services were terminated and post
permanency planning services were initiated with a permanent plan of long-term foster
care. At this time, Nancy's location was unknown and she had not complied with her

service plan. Samuel was still placed with his maternal grandmother and appeared to be
doing well.

Samuel remained in out-ofhome care for the next 11 years. When he was in the
4 th grade he was diagnosed with a learning disability and began receiving special
education services. By the 6th grade he was getting straight A's and he maintained his
educational success throughout much of his education. At age 11, Samuel left his

grandmother's home due to her increasing age and he began living with a maternal aunt.
Samuel experienced some behavioral problems during this transition. At age 15,
Samuel's mother began visiting with him sporadically. Samuel emancipatedfrom the
CWS at age 18. He spent a total of 13 years in the CWS.
Family Maintenance Case

Angela was a 5-year-old African American female who entered the CWS as a
result ofgeneral neglect by her mother. The maltreatment was initially reported by a
police officer. She had one prior referral to the CWS for which she received voluntary
family maintenance services. Angela had five siblings---one of whom was also in the

cws.

Angela was born in San Jose, California and was in kindergarten at the time she
entered the CWS. She had been living with her mother and two siblings.
Cynthia, Angela's mother, was a 28-year-old African American female. She was
born in San Francisco, California. She was employed in customer service; she had a high
school diploma and had attended some college. Cynthia was separated from Angela's
father, and was Angela's primary caretaker. Cynthia had a criminal history that included:
two arrests and one conviction for petty theft. Cynthia also reported that while married to
Angela's father, the couple experienced domestic violence.

30

Duane, Angela's father, was a 36-year-old African American male. He was born
in New Orleans, Louisiana. He was employed as a security guard; he had graduated from
high school and attended some college. He did not have stable housing at the time of the
incident and was staying with friends. Duane had a criminal history that included five
vehicle violations, two assault and battery charges, one charge for driving under the
influence of alcohol and drugs, and five charges of failing to pay court fines.
The Jurisdictional/Dispositional hearing was held approximately one month after
Angela first entered the CWS. At this hearing, Angela was placed with her mother in
family maintenance services. Both Angela's mother and father were ordered to attend
domestic violence counseling services. Cynthia was also ordered to attend a parenting
class, and Duane was allowed I-hour supervised visits twice a week.
At a 3-month interim hearing, Angela was attending counseling and her mother
was attending parenting classes. Three months later, at her 6-month review hearing,
Angela's case was dismissed and she remained with her mother. Angela spent a total of7
months in the CWS.

White
Family Reunification Case
David was a 14-year-old White male who entered the CWS as a result ofgeneral
neglect and emotional abuse by his father and mother. David's mother, who told hospital
emergency room staffthat she was unable to handle David's behavior, initially reported
the ma/treatment. At the time of this incident, David had 13 prior referrals to the CWS
and had received CWS services on one prior occasion. David had one brother and one
sister-neither of whom was in the CWS.
David was born and raised in California. At the time he entered the CWS, he had
been expelled from middle school and was not attending school. He also had behavior
problems including sexual acting out, destructive and aggressive behavior. He was
described as emotionally needy. David was in the mental health system, and he was
receiving counseling at the time of the incident. David had also been arrested on one prior
occasion for stealing and was not convicted for this offense. David and his father had
been living alone together when David entered the CWS.
Marianne, David's mother, was a 37-year-old White female. Marianne and
David's father were separated. Marianne was employed as a clerk in a grocery store and
was described as having a history of alcohol abuse problems. She also had a history of
depression. She was living with her boyfriend at the time of the incident and was
reportedly physically abusive toward her boyfriend.
Richard, David's father, was a 39-year-old White male. At the time of the incident
he was employed as a school bus driver and was described as having a problem with
alcohol abuse.

31

A Jurisdictional/Dispositional hearing was held approximately one month after
the initial report of the incident. At this hearing, David was placed in a level 14
community care facility and his case was assigned to family reunification services with
his mother and father. David was ordered to attend individual counseling and family
counseling with his parents. Marianne and David were ordered to attend joint counseling
with their son and had weekly 2-hour unsupervised visits.
At the time ofDavid's first 6-month review hearing, he was living at juvenile hall
and was charged with two felonies-one for molest and one for molest by means offorce.
Prior to entering juvenile hall, David had been placed at a therapeutic community care
facility where he was noted to have oppositional behavior and had run away several
times. David also had a psychiatric medical evaluation, and was not recommended for
psychotropic medications. During this time, David's father refused to attend parenting
classes and visited David on an infrequent basis. A 30-day continuance was ordered in
order to assess the possibility of dismissing David's child welfare dependency and
transferring his dependency to the juvenile justice system.
At the 30-day continuance hearing, David had been placed in a residential
treatment facility and was under the supervision of the probation department. His "300
dependency" status within the CWS was changed to a "602,, status and David became a
dependent ofthe juvenile justice system. From the initia[ incident to the dismissal, David
spent approximately 9 months in the CWS, after which he was discharged to the juvenile
justice system.
Family Maintenance Case
Nathan was an 8-year-old White male who entered the CWS as a result ofgeneral
neglect and emotional abuse by his mother and father. The maltreatment was initially
reported by a social worker. At the time of this incident Nathan had five prior referrals to
the CWS and had received CWS on one prior occasion. Nathan had two siblings-one of
whom was in the CWS.
Nathan was born in Stockton, California and was raised in California. He was in
the third grade at the time he entered the CWS and had been diagnosed with a learning
disability for which he was receiving special education services through his school.
Additionally, Nathan had a speech problem for which he received speech therapy. He
was also diagnosed with ADHD and an adjustment disorder with disturbed conduct. He
was described as guarded and somewhat closed off; he demonstrated aggressive and
disruptive behavior both at school and at home. At the time Nathan entered the CWS, he
was living with his mother, his mother's boyfriend and one sibling.
Sally, Nathan's mother, was a 31-year-old White female. At the time of the
incident, she was divorced from Nathan's father and was Nathan's primary caretaker.
Sally was employed, however her occupation was not indicated. Sally and her boyfriend
were also expecting another child at the time Nathan entered the CWS.

32

George, Nathan's father was a 33-year-old White male. At the time of the incident
he was remarried and was living with his second wife and one child. George had no
criminal record. Although he still had legal custody ofNathan, George was not his son's
primary caretaker.
At the time Nathan entered the CWS, his parents had spent several years in a
bitter custody battle over him. George was also described as having a more physically
punitive parenting style than Sally. The conflicts resulting from the custody battle and the
different discipline practices were reportedly distressing to Nathan

A Jurisdictional/Dispositional hearing was held approximately one month after
the initial report of the incident. At this hearing, Nathan was placed with his father and
his case was assigned to family maintenance services with his father and to family
reunification services with his moiher. Nathan was ordered to attend weekly individual
counseling. Both ofNathan's parents were ordered to attend a parenting without violence
class, as well as individual counseling. Sally was also ordered a 2-hour supervised visit
with Nathan once a week.
At the time ofNathan's first 6-month review hearing, he continued to live with his
father. He was receiving Victim Witness funds and was attending weekly counseling.
George had completed a parenting class, an anger management class and counseling. A
new order for George to attend family counseling was issued at this hearing. During this
time, Sally had criminal charges brought against her (although the record does not
indicate what she was charged with), and these charges were subsequently dismissed
through a plea agreement. Sally had also given birth three months prior to this hearing.
Although she was enrolled in a parenting class, she was not attending; however, she was
attending individual counseling. Additionally, Nathan's schoolteacher had written letter
to the court stating that Nathan's contact with his mother was causing him to be
disruptive.

a

At the time ofNathan's final hearing-a 90-day review-he was experiencing
improvement in his behavior and mental health, and his counseling services were
reduced to every other week. Nathan had experienced one school suspension for
disruptive behavior. Sally had completed a parenting class, but was living in different
state at the time of this hearing. Nathan's case was dismissed at this hearing with legal
custody to both his mother andfather, and physical custody to his father. From the time
of the initial incident, to the dismissal, Nathan spent 9 months in the CWS, after which he
was discharged to his father.
Hispanic/Latino
Family Reunification Case

Antonio was a ]-½-year-old Latino who entered the CWS as a result ofhis two
older half-sisters being sexually abused by his father and neglected by his mother. The

33

police initially reported the maltreatment to the CWS. Antonio had three siblings-and
his two older half-sisters entered the CWS prior to Antonio's entrance. At the time
Antonio entered the CWS, he was living with his mother and one older brother.
Hilda, Antonio's mother, was a 26-year-old Latina. At the time of the incident,
she was Antonio's primary caretaker. Hilda was born in Mexico, and she had been living
in the United States for 9 years. She did not speak English and required a Spanish
translator in court proceedings. Hilda was single and had never been married; she was
unemployed with an elementary school education and she was receiving AFDC.
Joseph, Antonio's father, whose age was not indicated in the case record, was
Latino. At the time Antonio entered the CWS, Joseph had been convicted of sexual
molesting his two step-daughters, and he was incarcerated in a state prison in California.
He had only recently entered prison at the time his son entered the CWS.

The case record did not indicate the date ofAntonio 's Jurisdictional/
Dispositional hearing. At this hearing, Antonio was placed in a family foster home and
his case was assigned to family reunification services. Hilda was ordered to attend a
parenting class offered in Spanish, as well as individual and family counseling. Joseph
was ordered parenting classes offered in prison.
Case records were missing for the 7 years following Antonio's initial entrance
into the CWS. The first court hearing record available was a 6-month review, at which
time Antonio was 8½-years-old, and living with a foster family. Family reunification
services had been terminated and Antonio was in post permanency planning services. At
this time, Antonio was having some behavioral problems. He was attending group
counseling and he was also ordered to begin additional counseling at his school. He was
having monthly visits with his mother and his siblings. At this time, Antonio expressed
his wish to have his foster parents become his legal guardians and continue living with
them until he reached the age of 18-Antonio's mother agreed with this plan.
At the following 6-month review, Antonio's foster parents became his legal
guardians. Antonio continued to have regular visits with his mother and siblings, and he
also visited his father in prison on one occasion. His father was released from prison and
was subsequently deported to Mexico. His whereabouts at the time of this hearing were
unknown, although he reportedly did not object to the legal guardianship for Antonio.
During the next 12 months, Antonio's mother was arrested for domestic violence
against her boyfriend and was incarcerated for an unspecified amount of time.
Additionally, over this time period, Antonio experienced physical abuse by his legal
guardians and legal guardianship was subsequently terminated. As a result, Antonio was
placed atthe Children's Shelter.

At the following 6-month review hearing, Antonio had left the Children's Shelter
and had been placed in another foster home; his permanent placement plan was long
term foster care. At this time, Antonio requested to be returned to his mother. Three

34

months later, Antonio's mother filed a petition with the court to request that her son be
returned to her. At the time of her request she was employed full-time and had been
granted unsupervised, weekly overnight visitations.

At the next 6-month review hearing, Antonio's mother withdrew her petition to
have her son returned to her-reportedly because she hadfailed to complete her initial
service case plan. The next 6-month hearing, the last hearing/or which information was
available in this case, indicated that Antonio was 12½-years-old, he was still living in a
family foster home and his permanent plan remained long-term foster care. Available
records indicated that Antonio spent at least 11 years in out of home care-although his
total length of time in the CWS was likely longer, as the last record available suggested
that he would not be returned to his mother and instead would remain in long term foster
care.
Family Maintenance Case

Veronica was an 8-year-old Hispanic female who entered the CWS as a result of
general neglect and emotional abuse by her mother. Veronica's teacher initially reported
the maltreatment. She had one prior referral to the CWS, but did not enter the system for
that referral. At the time of this incident, Veronica had eight siblings-seven of whom
were also in the CWS.
Veronica was born in San Jose, California and was living with her mother, father
and siblings at the time she entered the CWS.
Maria, Veronica's mother, was a 28-year-old Mexican American female. She
spoke some English and also required a Spanish translator in court proceedings. Maria
was born in Mexico and moved to the United States shortly after her birth; at the time her
daughter entered the CWS, she had been living in California for 27 years. Maria had an
11 th grade education and was working at a health clinic. At the time of the incident, Maria
was married to Veronica's father.
Rafael, Veronica's father, was a Mexican American male whose age was not
indicated in the case record. Rafael was born in Mexico, he did not speak English and
required a Spanish translator in court proceedings. At the time his daughter entered the
CWS, Rafael was employed as a construction worker and had a 6th grade education.
Rafael reportedly had an alcohol abuse problem. Additionally, he was experiencing a
back injury at the time of the incident-for which he was receiving worker's
compensation benefits.
At the time she entered the CWS, Veronica's family had a long history of
conflicts. Her parents experienced frequent marital discord and Rafael had a history
(dating back to the beginning of the marriage) of physically abusing Maria in front of
their children.

35

A Jurisdictional/Dispositional hearing was held approximately two months after
the initial report of maltreatment. During this 2-month interim, Veronica was living at the
Children's Shelter and was experiencing chronic stress and depression. Supervised visits
from both Veronica's mother and father were occurring during this time.

At the Jurisdictional/Dispositional hearing Veronica was placed back with her
parents, and her case was assigned to family maintenance services for both her mother
and father. Veronica and both of her parents were ordered to attend counseling. Maria

and Rafael were also ordered to attenq parent education classes. Additionally, Maria was
ordered to attend domestic violence counseling, and Rafael was ordered to attend a 12step program, and an alcohol treatment program.

At the first 6-month review hearing, Veronica was attending counseling regularly.
She had also been determined to need special education services and was deemed ·
appropriate for Severely Emotionally Disturbed (SED) services. Maria and Rafael .had
attended their parent education classes sporadically and had not attended counseling.
Rafael was attending 12-step meetings. Family maintenance services were ordered for
another six months in order for Maria and Rafael to make progress on their case plan.

At the following 6-month hearing, Veronica had been placed in special education
services at her school and was making academic progress. Her attendance at counseling
was less consistent, although her therapist reported that she her symptoms ofdepression
were lessening. Maria and Rafael had completed a parent education class; Maria did not
attend counseling and Rafael continued to attended 12-step meetings sporadically. At this
hearing, Veronica's case was dismissed and she remained with her parents; she spent
just over I-year in the CWS.
Asian American/Pacific Islander
Family Reunification Case

Brian was an I I-year-old Vietnamese male who entered the CWS as a result of
general neglect and caretaker absence/incapacity by his mother. Brian's school
counselor initially reported the maltreatment. He had one prior referral to the CWS, for

which he did receive CWS services. Brian had one older sister who was in the CWS in a
different state.
Brian was born in Vietnam and had been living in the United States since the age
of two. He was in the 5th grade and had some behavior problems including stealing at
school. At the time he entered the CWS, Brian was living with his mother, although from
the ages of two to nine he had lived with an aunt.
Tina, Brian's mother, was a 38-year-old Vietnamese female. Tina spoke English,
although her primary language was Vietnamese. Tina was born in Vietnam, and she had
been living in the United States for nine years. She was employed as a waitress, and she
reportedly stated that she felt proud that she did not rely on the welfare system. Tina also

36

reportedly left Brian alone for long periods of time. She was suffering from a delusional
disorder; she was experiencing grandiose delusions, and demonstrating assaultive
behavior at the time her son entered the CWS. Tina was divorced from Brian's father,
and she was Brian's primary caretaker.
Dan, Brian's father, was a Vietnamese male whose age was not indicated in the
case record. At the time Brian entered the CWS, Dan was living in a different state,
although his exact whereabouts were unknown. Dan's criminal history included one
conviction for a domestic violence incident during which he shot his wife-Brian's
mother. He was incarcerated for an unspecified amount of time for this crime, but had
been released at the time his son entered the CWS.

During the three weeks between the initial incident and the Jurisdictional/
Dispositional hearing, Brian was placed in emergency satellite housing where he was
reported to have adjusted well. Tina moved from her home during this time, and did not
provide her new address to the court. She refused to provide Brian with his clothing and
belongings, and she did not attend the Jurisdictional/Dispositional hearing.

At the Jurisdictional/Dispositional hearing, Brian was placed in a family foster
home and his case was assigned to family reunification services. Brian and his mother
were ordered to attend family counseling together. Tina was also ordered to attend a

parenting program focusing on child development issues (conducted in Vietnamese), a
cultural awareness program to assist her in dealing with adjustment and adaptation issues
(also conducted in Vietnamese), and individual counseling. Additionally, Tina was
ordered to undergo a psychological evaluation, and to refrain from using alcohol. She
was ordered 2-hour supervised visits once a week. Brian's father was not offered
services, as his whereabouts were unknown at the time of the hearing.

During Brian's first 6-month review hearing, his mother's visitation rights were
revoked due to non-compliance with her service plan. Visitation rights were to be
restored after Tina began participating in her court-ordered service plan. At this time,

she was reportedly exhibiting paranoid, unpredictable, delusional, and ambivalent
behaviors, and she refused to provide the CWS with her address or phone number. Brian
was living with a foster family at this time. He was reported to be loyal and protective of
his mother, and he expressed his wish to return to her home.

At the subsequent 6-month review hearing, Brian was in the 6th grade. He was an
above average student and was reported to be healthy and active. He remained with the
same foster family and had one incident where he stole $25. 00 from his foster parents'
children. Brian also completed a 14-session youth leadership project organized by DFCS.
Brian's mother did not comply with her court-ordered family reunification services, and
she had refused to sign the service plan. She was reported to be moving frequently,
although she did begin having regular visits with Brian. At this time, family reunification
services were terminated and Brian entered post permanency planning services.

37

At his next 6-month review hearing, Brian's permanent placement plan was
determined to be long term foster care. At this time he was in the 7th grade and was on
the honor roll. He was still with the same foster family, and he expressed his desire to
stay in this placement until he emancipatedfrom the system. Yet two months later, Brian

was placed at the Children's Shelter due to conflicts with his foster family. He was
reportedly not following house rules, was stealing food and was reportedly a bad
influence on his foster parents' own children. At the next 6-month review hearing, Brian
had left the Children's Shelter; he was placed in a new foster home that appeared to be
stable and he was reportedly adjusting well to his new placement.

Brian remained in the same foster home over the next 12 months. His school

grades dropped to a 2.00 Grade Point Average (GPA), but then improved to a 3.00 GPA.
He also obtained a part-time job at a fast food restaurant. At this time, Brian's mother had
stopped visiting with him.
Over the next 12 months Brian turned 14-years-old. He had one incident of petty
theft for which he was placed on probation. Outside of this incident, he was reported to
be doing well at home and at school. He remained with the same foster family and had
maintained a 3.5 GPA. Brian's mother had not visited him during this time and her
whereabouts were unknown.
Brian continued to maintain a 3.5 GPA over the next 18 months and he also
became involved in sports, particularly gymnastics. He was reported to be physically
healthy and described his future plans as graduating from high school·and then attending
the Air Force Academy. He remained with the same foster family. During this time,
Brian's mother was located in a different state.

Brian turned 17 over the following 12 months. He remained in the same
placement with his foster family and also remained focused on his school studies. He

began playing basketball and football. Although Brian was referred to the Independent
Living Program (ILP), he refused to attend. During this time, his mother returned to San
Jose, and she began receiving mental health services. Brain visited with his mother on a
few occasions after she returned to San Jose.

Brian 's final two review hearings indicated that he graduated from high school
and was accepted to a 4-year university. He also applied for financial aid and a work
study program. He expressed a desire to major in business accounting. Brain also

obtained a car and was approved for Medi-cal. During this time, Brian was participating
minimally in ILP. Brian's mother started taking psychiatric medications for her mental
illness and Brian and his mother began visiting more frequently with one another. Brian's

case l_Vas dismissed after he reached 18 years ofage and emancipated from the system.
He spent a total of 7 years in the CWS.
Family Maintenance Case

38

Ronald was an 8-year-old Korean male who entered the CWS as a result of
serious physical harm committed by his father. Ronald's teacher initially reported the
maltreatment. Ronald had no prior referrals to the CWS. He had one sister, who also

entered the CWS at this time. Ronald was living with his mother, father and sister at the
time he entered the CWS.

Carolyn, Ronald's mother was a 39-year-old Korean female. She did not speak
English and required a Korean translator in court proceedings. She was married to
Ronald's father and shared parenting duties with him.
William, Ronald's father, was a 44-year-old Korean male. He did not speak
English and required a Korean translator in court proceedings. William was born in
Korea. His criminal history included one conviction for fighting with a deadly weapon
for which he received two years probation. In a separate incident, he was convicted of
fighting and received one year probation. He was also convicted of driving without a
license and received two years probation for this offense.
The Jurisdictional/Dispositional hearing was held approximately 3 weeks after
the initial incident. At this hearing, Ronald was placed with his mother on the condition
that his father move out of the house. His case was assigned to family maintenance with
his mother and family reunification with his father. Ronald and his parents were ordered

to attend family counseling. His parents were also ordered to attend parenting classes and
couples counseling. Home supervision services were ordered three times a week for
Carolyn and Ronald. Ronald's father was ordered supervised visits twice a week.
Four interim hearings were held over the next four months, during which William
was ordered to attend a 52-week domestic violence program and Ronald and his sister
were ordered to attend a domestic violence group/or teenagers. Carolyn also filed a

temporary restraining order against William. During this time, Ronald's sister entered the
Children's Shelter for unspecified reasons. It was also reported that Ronald's parents
began attending meetings of the Parents, Families and Friends of Lesbians and Gays
(PFLAG) organization, although the specific reasons for this were not indicated.
Over the following 12 months, Ronald had academic problems and failed six of
his classes. Ronald's father was reportedly not attending his domestic violence program.
Nonetheless, William's service plan was changed from family reunification to family
maintenance and he returned to live with his wife and Ronald.

An interim hearing was held one month later, during which the court approved a
three-week vacation to Korea for Ronald, his mother and his sister. During this hearing, a
psychological evaluation was ordered for the whole family.
During Ronald's final 6-month review hearing,family maintenance services were
terminated for his father and he subsequently moved out ofthe home. It was also reported
that Ronald was behind in his school credits. After spendingjust under two years in the
CWS, Ronald case was dismissed and he was discharged to the care of his mother.

39

VI. Overall Closed Case Sample: Results
CHILD DEMOGRAPHICS AND SYSTEM-RELATED
CHARACTERISTICS
In order to provide context for the analyses to follow, we have included summary
tables from Phase I (See Attachment 7). Phase l's analysis was based on a point-in-time
sample of all open cases active during December 2000. 4399 cases were open during that
time of which 2721 were in out of home placement. A majority of the cases were
Hispanic/Latino (52.2%). Most children were in permanent placement (44.4%) with
30.2% in family maintenance, 19.7% in family reunification, and 5.7% in emergency
response. Among the cases in out of home placement, a majority were Hispanic/Latino
(50.8%). Tables 1 through 4 in Attachment 7 contain key variables regarding those cases
analyzed in Phase l, which provide a context for the review of the closed cases in Phase
2. In Phase 2, the CWRT continued its investigation of the disproportionate
representation of children of color in the child welfare system by analyzing a sample of
cases closed during an 18-month period (January 2000 through June 2001). Key case
characteristics were analyzed, including ethnic comparisons.
This section provides a description of ethnic differences in the demographic and
system-related characteristics for children and youth in the overall closed case sample
(N=l720).
The Sample
From 6761 total case closures over an 18-month period between January 2000 and
June 2001, a data file containing 1753 cases representing one child per family and one
case opening was constructed. From this total, we obtained valid information on I 720
cases from CWS/CMS.
Analyses
A series of exploratory bivariate analyses were conducted with child's ethnicity as
the main distinguishing variable. The purpose of this initial set of analyses was to identify
case differences related to ethnicity in order to help describe the experiences of each
ethnic group. The following sections will describe (1) general characteristics by ethnic
group, and (2) significant differences related to ethnic group.
Definition of Terms
Episode - An episode is considered the most recent case experience recorded in Santa
Clara County's CWS.
Other category - Those in the Other ethnic subgroup included Native Americans,
Ethiopians, White-Armenians, and White-Middle Easterners

40

Basic Demographics

Child Characteristics

The ethnicity of the sample was 11.1 % African American (190 of 1711 cases),
29.5% White (505 of 1711 cases), 46.4% Hispanic/Latino (794 of 1720 cases), 9.3%
Asian American/Pacific Islander (159 of 1711 cases), and 3.7% Other (63 of 1711 cases).
There was a significant relationship between gender and ethnicity. Overall, 53.2% of the
sample was female (910 of 1710 cases). The highest percentage of females was found in
the Hispanic/Latino ethnic group (56.7%, 450 of 1710 cases) and the highest percentage
of males was found in the Other ethnic group (60.3%, 38 of 63 cases). There was not a
significant relationship between ethnicity and age. The average age of the sample was
5.99 years (sd = 5.14). Asian American/Pacific Islander children were oldest, with an
average age of 6.78 years (sd = 5.29), and African American children were the youngest
with an average age of 5.45 years (sd = 5.05). Please see Table 1.
Table 1: Basic Demographics by Ethnic Group

N Size
Gendera
Female
Male

Age in years 0
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

1711

190
(11.1%)

505
(29.5%)

794
(46.4%)

159
(9.3%)

63
(3.7%)

910
(53.2%)

94
(49.5%)

259
(51.3%)

450
(56.7%)

82
(51.6%)

25
(39.7%)

800
(46.8%)

96
(50.5%)

246
(48.7%)

343
(43.3%)

77
(48.4%)

38
(60.3%)

5.99.
(5.14)

5.45
(5.05)

6.10
(5.07)

5.84
(5.11)

6.78
(5.29)

(5.85)

a Based on 1710 cases with valid information

b Based

on 1711 cases with valid information

Child's Language
There was a significant relationship between child's language and ethnicity. The
· majority of the sample spoke English (86.0%, 1453 of 1690 cases). Asian
American/Pacific Islander children were the most likely to speak an Asian/Pacific
Islander language (38.4%, 61 of 159 cases), and Hispanic/Latino children were the most
likely to speak Spanish (20.6%, 161 of780 cases). Please see Table 2.

6.15

41

Table 2. Child's Languagea
Total

Black

White

Latino

Asian/PI

Other

English

1453
(86.0%)

188
(99.5%)

497
(99.4%)

618
(79.2%)

94
(59.1%)

56
(90.3%)

Asian/PI
Language

64
(3.8%)

0
(0.0%)

1
(0.2%)

0
(0.0%)

61
(38.4%)

2
(3.2%)

Spanish

163
(9.6%)

0
(0.0%)

0
(0.0%)

161
(20.6%)

0
(0.0%)

2
(3.2%)

10
· (0.6%)

1
(0.5%)

2
(0.4%)

1
(0.1%)

4
(2.5%)

2
(3.2%)

Other

a

Based on 1690 cases with valid information

System-Related Factors
Voluntary Status
There was a significant relationship between voluntary status and ethnicity. A
slight majority of the whole sample was in involuntary services (56.6%, 933 of 1648
cases). Asian American/Pacific Islander children were most likely to be in voluntary
services (67.7%, 107 of 158 cases), while Other children were most likely to be in
involuntary services (80.6%, 50 of 62 cases). Among the four ethnic groups, Whites were
most likely to be in involuntary services. Please see Table 3.
Table 3: Voluntary Status8

Voluntary
Not
Voluntary

Total

Black

White

Latino

Asian/PI

Other

715
(43.4%)

71
(39.2%)

174
(36.6%)

351
(45.5%)

107
(67.7%)

12
(19.4%)

933
(56.6%)

110
(60.8%)

302
(63.4%)

420
(54.5%)

51
(32.3%)

50
(80.6%)

a Based on 1648 cases with valid information

42

Reason for Removal
There was a significant relationship between the reason for removal and ethnicity.
White children had the highest percentage of other maltreatment (43.0%, 167 of 388
cases), the most common reason for removal. Hispanic/Latino children had the highest
percentage of general neglect (15.9%, 92 of 578 cases). Asian/Pacific children (17.4%,
20 of 115 cases), and, African American children( 16 .1 %, 26 out of 161 ), had somewhat
higher percentages of severe neglect than other groups. Asian American/Pacific Islander
children had a very high rate of physical abuse (42.6%, 49 of 115 cases). Please see Table

4.

Table 4: Reason for Removala
Total

Black

White

Latino

Asian/PI

Other

General
Neglect

176
(13.5%)

20
(12.4%)

52
(13.4%)

92
(15.9%)

6
(5.2%)

6
(10.5%)

Severe
Neglect

184
(14.2%)

26
(16.1%)

48
(12.4%)

83
(14.4%)

20
(17.4%)

7
(12.3%)

Physical
Abuse

328
(25.3%)

41
(25.5%)

83
(21.4%)

141
(24.4%)

49
(42.6%)

14
(24.6%)

Sexual Abuse

80
(6.2%)

7
(4.3%)

26
(6.7%)

43
(7.4%)

3
(2.6%)

1
(1.8%)

Emotional
Abuse

45
(3.5%)

7
(4.3%)

12
(3.1%)

15
(2.6%)

6
(5.2%)

5

(8.8%)

Other
Maltreatment

486
(37.4%)

60
(37.3%)

167
(43.0%)

204
(35.3%)

31
(27.0%)

24
(42.1%)

a Based

on 1299 cases with valid information

Initial Out-of-Home Placement Facility Type
There was a significant relationship between initial out-of-home placement
facility type and ethnicity. Asian/Pacific children were far more likely than other children
to be placed at the most common initial out-of-home placement type, Children's Shelter

43

(47.8%, 55 of 115 cases), and less likely at a relative's home (14.8%, 17 out of 115).
White children (8.2%, 32 of388) and Other children (7.0%, 4 out of 57) were more likely
than others to be placed at a group home, and Hispanic/Latino children at a relative home
(31.7%, 183 of 578 cases). Please see Table 5.
Table 5: Initial Out of Home Placement Facility Typea
Total

Black

White

Latino

Asian/PI

Other

Foster Family
Home

278
(21.4%)

38
(23.6%)

83
(21.4%)

118
(20.4%)

25
(21.7%)

14
(24.6%)

Group Home

70
(5.4%)

6
(3.7%)

32
(8.2%)

23
(4.0%)

(4.3%)

5

4
(7.0%)

390
(30.0%)

49
(30.4%)

99
(25.5%)

172
(29.8%)

(47.8%)

55

15
(26.3%)

Relative Home

367
(28.3%)

43
(26.7%)

108
(27.8%)

183
(31.7%)

17
(14.8%)

16
(28.1%)

Foster Family
Agency

138
(10.6%)

18
(11.2%)

47
(12.1%)

58
(10.0%)

11
(9.6%)

4
(7.0%)

Guardian
Home/ Court
Specified
Home

56
(4.3%)

7
(4.3%)

19
(4.9%)

24
(4.2%)

2
(1.7%)

4
(7.0%)

Children's
Shelter/
Receiving
Home/NonEA/AFDC

a

Based on 1711 cases with valid information

Number of Times Removed from Family in Current Episode
There was a significant relationship between the number of times a child was
removed from their family in current episode and ethnicity. On average, children in the
sample were removed 1.36 times (sd = 0.71). Other children had the highest average
number of times removed from their family with 1.54 (sd = 0.87), and Asian
American/Pacific Islander children had the lowest average with 1.16 times (sd = 0.47).
Please see Table 6.

44

Table 6. Number of Times Removed from Family in Current Episodea

Average
(sd)
a

Total

Black

White

Latino

Asian/PI

Other

1.36
(0.71)

1.47
(0.76)

1.34
(0.78)

1.35
(0.66)

1.16
(0.47)

1.54
(0.87)

Based on 1331 cases with valid information

Number of Unique Placement Homes in Current Episode
The.re was a significant relationship between the number of unique placement
homes in current episode and ethnicity. The sample had an average of 3.56 (sd = 3.82)
unique placement homes in current episode. Other children had the highest average
number of unique placement homes in current episode with 4.09 (sd =:= 4.21), and Asian
American/Pacific Islander children had the fewest average number of unique placement
homes in current episode with 2.26 (sd = 2.17). Please see Table 7.
Table 7. Number of Unique Placement Homes in Current Episodea

Average
(sd)
a Based

Total

Black

White

Latino

Asian/Pl

Other

3.56
(3.82)

3.71
(4J7)

3.76
(3.82)

3.59
(3.84)

2.26
(2.17)

4.09
(4.21)

on 1331 cases with valid information

Average Stay (in days) per Placement Facility in Current Episode
There was a significant relationship between the average stay (in days) per
placement facility in current episode and ethnicity. For the whole sample, the average
number of days a child stayed in one placement in their current episode was 241. 73 (sd =
485.74).African American children had the lengthiest average stay per placement in
current episode with 376.60 days (sd = 780.24), and Other children had the shortest
average stay per placement in current episode with 197.65 days (sd = 254.55). Please see
Table 8.
Table 8. Average stay (in days) per Placement Facility in Current Episodea

Average
(sd)
a Based

Total

Black

White

Latino

Asian/Pl

Other

241.73
(485.74)

376.60
(780.24)

198.61
(304.70)

245.16
(455.29)

205.68
(639.28)

197.65
(254.55)

on 1331 cases with valid information

45

Number of Placements in Current Episode
There was a significant relationship between· the number of placements in episode
and ethnicity. On average, children in the sample had 2.83 placements (sd = 2.90). White
children had the highest average number of placements in current episode with 3.11 (sd=
3 .20), and Asian American/Pacific Islander children had the fewest average number of
placements in current episode with 2.10 (sd = 2.09). Please see Table 9.
Table 9: Number of Placements a

Average
(sd)
8

Total

Black

White

Latino

Asian/PI

Other

2.83
(2.90)

2.73
(2.97)

3.11
(3.20)

2.80
(2.80)

2.10
(2.09)

3.05
(2.65)

8ased on 1299 cases with valid infonnation

Total Length of Time (in months) in Out of Home Placement in Current Episode
There was not a significant relationship between the total length of time (in
months) in out-of-home placement in current episode and ethnicity. On average, children
in the sample spent 12.92 months (sd = 19.94) in out-of-home placement. African
American children spent the lengthiest average time in out-of-home placement with 15.13
months (sd = 25.93), and Asian American/Pacific Islander children had the shortest
average time in out-of-home placement with 10.85 months (sd = 24.92). Please see Table
10.
Table 10. Total length of time (in months) in Out of Home Placement in Current Episodea

Average
(sd)
a Based

Total

Black

White

Latino

Asian/Pl

Other

12.92
(19.94)

15.13
(25.93)

11.85
(14.94)

13.46
(20.04)

10.85
(24.92)

12.59
(17.81)

on 1292 cases with valid information

Number of Episodes
There was not a significant relationship between the number of episodes and
ethnicity. Children in the sample had an average of 1.19 (sd = 0.48) episodes. Other
children had the highest average number of episodes (1.29, sd = 0.52), and Asian
American/Pacific Islander had the lowest average number of episodes ( 1.11, sd = 0.33).
Please see Table 11.

46

Table 11. Number of Episodesa

Average
(sd)

a

Total

Black

White

Latino

Asian/Pl

Other

1.19
(0.48)

1.22
(0.47)

1.19
(0.41)

1.20
(0.53)

l.ll
(0.33)

1.29
(0.52)

Based on 1648 cases with valid information

Total Length of ca·se (in years)
There was a significant relationship between the total length of the case (in years)
and ethnicity. On average, children in the sample had a case length of 1.85 years (sd =
2. 74). African American children had the longest average length of case with 2.67 years
(sd = 3.66), and Asian American/Pacific Islander children had the shortest average length
of case with 1.26 years (sd = 2.38). Please see Table 12.
·
Table 12: Total Length of Case (in yearst

Average
(sd)
8

Total

Black

White

Latino

Asian/Pl

Other

1.85
(2.74)

2.67
(3.66)

1.84
(2.49)

1.76
(2.69)

1.26
(2.38)

2.15
(2.52)

Based on 1648 cases with valid information

Total Number of Workers Assigned to Case over Time in the Current Episode
There was a significant relationship between the total number of workers assigned
to case over time in current episode and ethnicity. Children in the sample had an average
of 4.90 workers (sd = 2.74). White children had the highest average number of workers
with 5.11 (sd = 2.92), and Asian American/Pacific Islander children had the fewest
average number of workers with 4.35 (sd = 1.94). Please see Table 13.
Table 13: Total Number of Workers Assigned to Case over Time in the Current Episodea

Average
(sd)
a Based

Total

Black

White

Latino

Asian/PI

Other

4.90
(2.74)

4.90
(2.89)

5.11
(2.92)

4.90
(2.76)

4.35
(1.94)

4.60
(2.11)

on 1671 cases with valid information

47

Last Out-of-Home Placement Type
There was not a significant relationship between last placement type and
ethnicity. In the sample, Hispanic/Latino children were more likely than others to be
placed at the most common placement type, a relative home (39.2%, 176 of 449 cases).
Asian American/Pacific Islander children were especially likely to be placed at a foster
family home (40.3%, 25 of62 cases) or foster family agency (19.4%, 12 out of 62).
African American (IO .3 %, 12 out of 117) and White (9. 7%, 31 out of 319) children were
more likely than others to be placed at a group home, and Other children at a children's
shelter (15.9%, 7 out of 44). Please see Table 14.

48

Table 14: Last Out-of-Home Placement Typea
Total

Black

White

Latino

Asian/Pl

Other

Foster Family
Home

270
(27.2%)

36
(30.8%)

87
(27.3%)

108
(24.1%)

25
(40.3%)

14
(31.8%)

Group Home

76
(7.7%)

12
(10.3%)

31
(9.7%)

26
(5.8%)

4
(6.5%)

3
(6.8%)

Children's
Shelter

109
(11.0%)

10
(8.5%)

32
(10.0%)

56
(12.5%)

4
(6.5%)

7
(15.9%)

Relative
Home

344
(34.7%)

35
(29.9%)

106
(33.2%)

176
(39.2%)

14
(22.6%)

13
(29.5%)

Foster Family
Agency

142
(14.3%)

19
(16.2%)

47
(14.7%)

60
(13.4%)

12
(19.4%)

4
(9.1%)

Guardian
Home/ Court
Specified
Home

50
(5.0%)

5
(4.3%)

16
(5.0%)

23
(5.1%)

3
(4.8%)

3
(6.8%)

8

Based on 911 cases with valid information

Age at Time of Case Closure for the Current Episode

There was not a significant relationship between age at time of case closure for
current episode and ethnicity. On average, children in the sample were 8.16 years-of-age
(sd = 5.72) at the time of case closure. Other children were the oldest (on average) at time
of case closure with an age of9.36 years (sd = 6.39), and African American children
were youngest (on average) at time of case closure with an age of7.86 years (sd = 5.60).
Please see Table 15.
Table 15. Age at Time of Case Closure for the Current Episode8

Average
(sd)
a Based

Total

Black

White

Latino

Asian/PI

Other

8.16
(5.72)

7.86
(5.60)

8.30
(5.79)

8.00
(5.61)

8.45
(5.81)

9.36
(6.39)

on 1646 cases with valid information

49

Service Type at Case Closure
There was a significant relationship between service type at closure of case and
ethnicity. Overall, the majority of cases ended in family maintenance (60.3%, 1031 of
1711 cases). Amongst the services types at closure, Other children had the highest
percentage of emergency response service at closure (4.8%, 3 of 63 cases), Asian
American/Pacific Islander children had the highest percentage of family maintenance
(76.1 %, 121 of 159 cases), Other children had the highest percentage of family
reunification (7.9%, 5 of 63 cases) and African American children had the highest
percentage of permanent placement (40.5%, 77 of 190 cases). Please see Tablel6.
Table 16: Service Type at Case Closurea
Total

Black

White

Latino

Asian/PI

Other

Emergency
Response

46
(2.7%)

7
(3.7%)

14
(2.8%)

19
(2.4%)

3
(1.9%)

3
(4.8%)

Family
Maintenance

1031
(60.3%)

103
(54.2%)

288
(57.0%)

485
(61.1%)

121
(76.1%)

34
(54.0%)

Family
Reunification

81
(4.7%)

3
(1.6%)

31
(6.1%)

40
(5.0%)

2
(1.3%)

5
(7.9%)

Permanent
Placement

553
(32.3%)

77
(40.5%)

172
(34.1%)

250
(31.5%)

33
(20.8%)

21
(33.3%)

a Based on 1711 cases with valid information

Case Closure Type
There was a significant relationship between case closure type and ethnicity.
Asian American/Pacific Islander children had the highest percentage of the most
frequently occurring case closure type, court ordered termination. (47.5%, 75 out of 158
cases). Other children had the ·highest percentage of being reunified with parent or
guardian (11.5%, 7 of61 cases) and emancipation (13.1%, 8 of61 cases). White children
had the highest percentage of adoption (18.5%, 88 of 475). Hispanic/Latino children had
the highest percentage of family stabilization (34.9%, 267 of764 cases). Please see Table
17.

50

Table 17: Case Closure Type 8
Total

Black

White

Latino

Asian/Pl

Other

Adoption

261
(15.9%)

30
(16.8%)

88
(18.5%)

118
(15.4%)

19
(12.0%)

6
(9.8%)

Emancipation

96
(5.9%)

14
(7.8%)

24
(5.1%)

43
(5.6%)

7
(4.4%)

8
(13.1%)

Family
Stabilized
(FM)

523
(31.9%)

55
(30.7%)

140
(29.5%)

267
(34.9%)

48
(30.4%)

13
(21.3%)

Guardianship
established or
placement
with relative

108
(6.6%)

14
(7.8%)

35
(7.4%)

52
(6.8%)

4
(2.5%)

3
(4.9%)

Reunified with
parent or
guardian,
court or noncourt specified

66
(4.0%)

8
(4.5%)

27
(5.7%)

20
(2.6%)

4
(2.5%)

7
(11.5%)

Incarceration,
runaway, or
medical
services

49
(3.0%)

10
(5.6%)

17
(3.6%)

18
(2.4%)

1
(0.6%)

3
(4.9%)

Court ordered
termination
and other

534
(32.6%)

48
(26.8%)

144
(30.3%)

246
(32.2%)

75
(47.5%)

21
(34.4%)

a

Based on 1637 cases with valid information

51

Summary of Key Findings from the Overall Closed Case Sample
An analysis ofthe sample (N= 1720) ofcases closed in an 18-month period
between January 2000 and June 2001 reveal that children in the various ethnic groups
tend to be significantly different from one another in several areas.
Basic Demographics
In descending order of representation, the ethnic breakdown of children in the
closed case sample for the specified 18-month period was: Hispanic/Latino children
(46.4%), White children (29 .5%), African American children ( 11.1 %), Asian
American/Pacific Islander children (9.1%), and Other children (3.7%). The order of
representation reflects the order of representation in the county's CWS. There were
significantly more females than males in the sample, and Hispanic/Latino children had a
higher than average number of females. The average age of the sample was
approximately 6 years of age. While ethnic differences in age were not significant,
African American children tended to be younger than the average, while White, Asian
American/Pacific Islander and Other children tended to be older than the average.
Additionally, Asian American/Pacific Islander children (38.4%) and Hispanic/Latino
children (20 .6%) had the highest rates of being non-English speaking.
System-Related Factors
There were significant differences among the ethnic groups in their likelihood of
being in voluntary versus involuntary services. In general, African American children
(39.2%), White children (36.6%), and Other children (19.4%) had lower than average
rates ofvoluntary services. Asian American/Pacific Islander children had the highest
rates ofvoluntary family maintenances services. There were also significant differences
related to the reason for removal. Hispanic/Latino children had the highest percentage of
general neglect (15.9%, 92 of 578 cases). Asian American/Pacific Islander children had
the highest rate ofphysical abuse than any other ethnic group (42.6%) and they also had
a higher than average rate of severe neglect (17.4%).
Significant differences were also found in the types of initial out-of-home
placements. Asian American/Pacific Islander children had the highest rate of initially
being placed at the most common initial first placement, the Children's Shelter (47. 8%),
and were less likely to be placed in a relative 's home. White children were most likely to
be initially placed in a relative home and also had the highest rate of being placed at a
Group Home (8.2%). After the Children 's Shelter, the most common initial first
placement for African American and Hispanic/Latino children was a relative home. The
number of times a child is removed from their family in the current episode was also
significantly different among the ethnic groups. African American (1.47) and Other
(1.54) children had the highest average number of times removed from family in current
·
episode.

52

Experiences in out-of-home placement were significantly different among the
ethnic groups. African American children had a comparatively high average of unique
placement homes (3. 71) and a high average number ofdays in each out-ofhome
placement (376.60)-afinding that is consistent with African American children's higher
than average total length of case (2.67 years). Hispanic/Latino children (3. 76) also
experienced a relatively high number of unique placement homes-although they had
lower than average stays per placement, suggesting that these children have multiple
placements and relatively short stays in each placement. Asian American/Pacific Islander
(1.26 years), and Hispanic/Latino children (1. 76) had lower than average total case
lengths. Additionally, White children tended to have a higher than average number of
workers assigned to their case over time (5.11), and Asian American/Pacific Islanders
tended to have a lower than average number ofworkers (4.34).
At case closure, significant differences were also found in the sample. African
American children had the highest rate ofpermanent placement (40.5%) as their last
service type-perhaps suggesting that once removed from their home, African American
children are not likely to return. Asian American/Pacific Islander children had the
highest rate offamily maintenance (76.1%) at case closure-and the lowest rate of
permanent placement (20.8%)'. The case closure type was also significantly different
among the ethnic groups. White children had the highest rate ofadoption (18.5%).
African American (7.8%) and Other children (13. 1%) had higher than average rates of
emancipation. Hispanic/Latino children had a relatively high rate offamily stabilization
(34.9%), yet a relatively low rate offamily reunification (2.6%)-suggesting that
perhaps, like African American children, once Hispanic/Latino children are removed
from the home, they are not likely to return. Last, African American children had the
highest rate of incarceration, runaway or medical services (5. 6%).
While this description provides important information, we are unable to examine
variables that are not included in the CWS/CMS database. Information about the child
and family not contained in CWS/CMS include the child's history in the CWS, family
characteristics, and the impact of system-related factors on services. In order to augment
the information provided in CWS/CMS, we conducted in-depth and extensive case record
review of 403 child welfare cases, the results of which are provided in the following
section.

53

VII. Case Record Reviews
One of the primary objectives of Phase 2 of this study was to examine pathways
through the CWS and explore ways in which pathways might differ for different
racial/ethnic groups. Please see Figure I for a flow chart of possible pathways and key
choice points in Santa Clara county's CWS. The following section describes findings
based on 403 child welfare case records. The case record review enabled us to gather
extensive information on individual child characteristics, family related characteristics,
services and system related changes - inf<;>rmation that was not available through the
analysis of CWS/CMS data that was presented in the preceding section.
The first part of this section presents ethnic differences in child and family related
characteristics. Part 2 describes system related characteristics and differences by
ethnicity. Ethnic differences in services and other interventions ordered at the
jurisdictional/dispositional hearing are contained in part 3 of this section, and ethnic
differences in the number of hearings and case changes as children progress through the
system are presented in part 4.

The Sample
A combination of random stratified and purposive sampling was used to select a
set of 403 cases from the main sample of child welfare cases (N = 1720) closed during an
18 month period that ran from January 2000 and June 2001. The criteria for stratification
was ethnicity (recoded into 5 subgroups: African American, Hispanic/Latino, White,
Asian American/Pacific Islander, and Other), service type identified at the initiation of
the case (either ER, PP, or FR versus FM), and age group (0 to 6 years, 7 to 13 years, and
older than 13 years old). These strata were created in order to obtain as representative a
sample as possible reflecting the cases managed by Santa Clara County's CWS. This
subsample of 403 cases excludes voluntary family maintenance cases (see Section IV for
a complete description of sampling methods used).
We were unable to meet the target sample size through true random stratification
due to a large number of missing case files across strata. Approximately two-thirds of the
way through the sampling and data extraction phases, cases within strata with few
children (i.e., Asian American/Pacific Islanders and African Americans) had to be
selected purposefully in order to reach the minimum 400 caseload sample and improve
representation.

Analyses
A series of exploratory bivariate analyses were conducted with child's ethnicity as
the main distinguishing variable. The purpose of this initial set of analyses was to identify
case differences related to ethnicity to help describe the experiences of each ethnic group.
Key variables, such as mental illness, welfare eligibility, substance abuse, and
others were coded using a forced-choice "yes," "no," or "not indicated." A "yes" appears

54

in the following tables as "indicated" and a "no" or "not indicated" appears as "not
indicated in the following tables. If a variable is not indicated that does not mean it did
not occur-but rather means that it did not appear in the court records.
Definition of Terms

Episode - An episode is considered the most recent case experience recorded in Santa
Clara County's CWS.
Other category - Those in the Other ethnic subgroup included Native Americans,
Ethiopians, White-Armenians, and White-Middle Easterners.

55

Figure 1: Possible Pathways and Key Choice Points In the Child Welfare System
Referral
(Report of Abuse or Neglect)

Emergency Response
(Screening)

Case Closed

Case Opened:
Further
Investigation

Jurisdictional Hearing:
Decision Regarding Substantiation
of Abuse/Neglect

Case Dismissed:
Unsubstantiated
Allegations

Family
Maintenance

Case Opened:
Voluntary Family
Maintenance

Case Closed:
Short-Term Services
Provided and Family
Stabilized Within 6
Months

Dispositional Hearing:
Substantiated Allegations,
Further Action and Services

Family Reunification:
Placement in FFH, FFA, Group
Home, Kincare, or Other Facility

Bypass

Hearings: Including 6-Month Review

Case Closed:
Family Stabilized
Through Family
Maintenance

Case Closed:
Family Stabilized
Through Family
Reunification

Case Closed:
Permanent Placement into
Adoption, Guardianship,
or Long Term Care

56

1. General Child and Family Characteristics By Ethnic Group
Results
All data in this section were obtained from the case record extraction database.

Child Characteristics
Basic Demographics
The case record review sample included: 50 African Americans (12.4%), 136
Whites (33.7%), 141 Hispanics/Latinos (35.0%), 31 Asian American/Pacific Islanders
(7.7%), and 45 Others (11.2%). Sex of the child was identified in 401 of the 403 cases:
219 (54.6%) were female and 182 (45.4%) were male.
The children's average age was 7.50 years (sd = 5.03). There was a significant
difference in the child's age by ethnic group. African American children were an average
of 5.41 years old (sd = 4.93), Asian American/Pacific Islanders 6.64 years old (sd =
4.46), Hispanics/Latinos 7.83 years (sd = 4.94), Whites 7.93 years (sd = 5.02) and Others
8.02 years (sd = 5.40). Please see Table 18 for the basic demographics by ethnic group.
Table 18: Basic Demographics by Ethnic Group

N Size
Gendera
Female
Male

Age in yearsb
Avg.
(sd)
a

b

Total

Black

White

Latino

Asian/PI

Other

403

50
(12.4%)

136
(33.7%)

141
(35.0%)

31
(7.7%)

45
(11.2%)

219
(54.6%)

26
(52.0%)

67
(49.6%)

89
(63.1%)

17
(54.8%)

20
(45.5%)

182
(45.4%)

24
(48.0%)

68
(50.4%)

52
(36.9%)

14
(45.2%)

24
(54.5%)

7.51
(5.03)

5.41
(4.93)

7.93
(5.02)

7.83
(4.94)

6.64
(4.46)

8.02
(5.40)

Based on 401 cases with valid information
Based on 386 cases with valid information

57

Child's Language
There was a significant relationship between child's language and ethnicity. The
majority of the sample spoke English (87.6%, 346 of395 cases). Among the various
languages, 54.8% of Asian American/Pacific Islander children (17 of 31 cases) spoke
Asian/Pacific Islander languages and 17.6% of Hispanic/Latino children spoke Spanish
(24 of 136 cases). Please see Table 19.
Table 19: Child's Languagea
Total

Black

White

Latino

Asian/PI

Other

English

346
(87.6%)

50
(100.0%)

133
(99.3%)

111
(81.6%)

13
(41.9%)

39
(88.6%)

Asian/Pl
Language

19
(4.8%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

17
(54.8%)

2
(4.5%)

Spanish

26
(6.6%)

0
(0.0%)

0
(0.0%)

24
(17.6%)

0
(0.0%)

2
(4.5%)

Other

4
(1.0%)

0
(0.0%)

1
(0.7%)

1
(0.7%)

1
(0.7%)

1
(0.7%)

a Based

on 395 cases with valid information.

Child's Need for Translator in Court Proceedings
There was a significant relationship between a child's need for a translator in
court proceedings and ethnicity. African American and White children had no need for
translators in court proceedings, given that almost all were born and raised in the United
States. However, 5.7% of Hispanic/Latino children (8 of 141 cases), 3.2% of Asian
American/Pacific Islanders (I of 31 cases), and 4.4% of Others (2 of 45 cases) needed
translators in court proceedings. This is consistent with the immigration experience in
that children from ethnic groups non-native to the United States are more likely to be less
skilled in English. Please see Table 20.

58

Table 20: Child's Need for Translator in Court Proceedings

Indicated
Not Indicated

Total

Black

White

Latino

Asian/Pl

Other

(2.7%)

11

0
(0.0%)

0
(0.0%)

8
(5.7%)

1
(3.2%)

2
(4.4%)

392
(97.3%)

(0.0%)

136
(0.0%)

133
(94.3%)

30
(96.8%)

43
(95.6%)

so

Child's Years of Education at Case Opening
There was no significant relationship between child's years of education and
ethnicity. Using kindergarten as the first year of education, the overall sample of children
had 2. 75 years of education. Hispanic/Latino children had the highest average number of
years of education, with 3 .14 years ( sd = 3 .86), and African American had the lowest
average number of years of education, with 1.67 years (sd = 3 .08). Please see Table 21.
Table 21: Child's Years of Education at Case- Openinga

Average
(sd)
a

Total

Black

White

Latino

Asian/PI

Other

2.75
(3.44)

1.67
(3.08)

2.93
(3.38)

3.00
(3.57)

2.28
(2.79)

3.14
(3.86)

Based on 316 cases with valid information

Child's Place of Birth
There was a significant relationship between a child's place of birth and ethnicity.
Similar to the child's need for translator in court proceedings results above, African
American and White children were generally native to the United States with almost no
immigrants. The Other group was also indigenous to the United States. Each of these
groups only had one immigrant. However, 7 .1 % of Hispanic/Latino children (10 of 141
cases) and 22.6% of Asian American/Pacific Islanders (7 of 31 cases) were born outside
of the United States. Please see Table 2

59

Table 22: Child's Place of Birth
Total

Black

White

Latino

Asian/Pl

Other

Indicated
Outside U.S.

20
(5.0%)

1
(2.0%)

1
(0.7%)

10
(7.1%)

7
(22.6%)

1
(2.2%)

Indicated U.S.
or Assumed
U.S.

383
(95.0%)

49
(98.0%)

136
(99.3%)

131
(92.9%)

24
(77.4%)

44
(97.8%)

Child Mental Illness
There was a significant relationship between identified child mental illness and
ethnicity. Percentages of children identified as having mental health problems were as
follows: African American 6.0% (3 of 50 cases), Whites 14.7% (20 of 136 cases),
Hispanics/Latinos 10.6% (15 of 141 cases), Asian American/Pacific Islanders 0.0% (no
cases), and Others 4.4% (2 of 45 cases). Please see Table 23.
Table 23: Child Mental Illness
Total

Black

White

Latino

Asian/PI

Other

Indicated

40
(9.9%)

3
(6.0%)

20
(14.7%)

15
(10.6%)

0
(0.0%)

2
(4.4%)

Not Indicated

363
(90.1%)

47
(94.0%)

116
(85.3%)

126
(89.4%)

31
(100.0%)

43
(95.6%)

Child Behavioral Problems
There was a significant relationship between child behavioral problems and
ethnicity. Within the African American group 18.0% (9 of 50 cases) of children were
indicated as having a behavioral problem, Whites 33.8% (46 of 136 cases),
Hispanics/Latinos 21.3% (30 of 141 cases), Asian American/Pacific Islanders 22.6% (7
of 31 cases), and Others 13.3% (6 of 45 cases). Please see Table 24.

60

Table 24: Child Behavioral Problems
Total

Black

White

Latino

Asian/PI

Other

Indicated

98
(24.3%)

9
(18.0%)

46
(33.8%)

30
(21.3%)

7
(22.6%)

6
(13.3%)

Not Indicated

305
(75.7%)

41
(82.0%)

90
(66.2%)

111
(78.7%)

24
(77.4%)

39
(86.7%)

61

Father Characteristics
Considerably more information was available in case records on mothers than was
available on fathers. At most, data were recorded for 52% of African Americans fathers,
61 % of Asian American/Pacific Islander, 69% of Hispanics/Latinos, and 71 % of Whites.
The limited information that could be found describing fathers reflects the fact
that in the present sample, fathers were less likely than mothers to be living with children
at the time of child welfare involvement. Thus, whereas 70.2% of children were reported
as living with their mothers, only 29.9% of children were reported as living with their
fathers at the time of the incident requiring child welfare intervention. At least 32% of
fathers, but only 9% of mothers, did not have legal custody of the child. Across all ethnic
groups, the vast majority of mothers were designated as primary caretaker (68.0% of
African American, 72.8% of White, 75.9% of Hispanics/Latino and 74.2% of Asian
American/Pacific Islander mothers). Likewise, in the majority of cases, the mother was
designated as having custody of the child (66.0% of African American, 71.3% of White,
70.2% of Hispanics/Latino and 68.9% of Asian American/Pacific Islander. Since mothers
were more closely involved in the CWS, child welfare workers had better access to the
experience of mothers. Consequently, it appeared that workers were more likely to report
pertinent information on mothers when they entered information into the case record.
The relative lack of information is more problematic still because it is
disproportionate across racial and ethnic groups. Thus, rates of missing information vary
from one group to another and are greatest for African Americans and least for Asian
American/Pacific Islanders.
This disproportionate loss of information compromises our ability to carry out
meaningful comparisons among racial and ethnic groups. That is because in groups
where information is more complete, there is more opportunity for events, when they
occur, to be recorded. As a result, reporting both of negative and positive characteristics
is exaggerated in groups with higher rates of reporting. Because of this imbalance,
descriptive data on fathers are less accurate than data on mothers.
Data on fathers have therefore not been reported in detail for present purposes.
Father data were in fact analyzed and for the most part confirmed what was reported as
characterizing children and mothers. Nevertheless detailed information on fathers is not
presented in this document. Father data are considered not to be sufficiently accurate for
detailed reporting and, in too many instances, to lend themselves to misinterpretation.

62

Mother Characteristics
Mother As Primary Caretaker and Custodian
The relationship between primary caretaker and having custody of the child and
ethnicity was not statistically significant. In the total sample, however, almost threequarters of the cases (73.2%) identified the mother as the primary caretaker and over twothirds (69. 7%) of mothers as having custody of her child. Please see Table 25.
Table 25: Mother as Primary Caretaker and Custodian

Mother as Primary
Caretaker
Mother Having
Custod of Child

Total

Black

White

Latino

Asian/PI

Other

295
73.2%
281
69.7%

34
68.0%
33
66.0%

99
72.8%
97
71.3%

107
75.9%
99
70.2%

23
74.2%
21
67.7%

32
71.1%
21
68.9%

Mother's Age
There was a significant difference in the mother's age at time of incident by
child's ethnic group. African American mothers were an average of 30.59 years old (sd =
6.87), Latinas 31.60 years (sd = 6.89), Asian American/Pacific Islanders 33.65 years old
(sd = 8.05), Whites 34.16 years (sd = 6.80) and Others 34.83 years (sd = 8.78). Please see
Table 26.
Table 26: Mother's Age at Time oflncidenta
Age in years
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

32.85
(7.30)

30.59
(6.87)

34.16
(6.80)

31.60
(6.89)

33.65
(8.05)

34.83
(8.78)

a Based on 362 cases with valid information

Mother's Need for Translator in.Court Proceedings
There was a significant relationship between a mother's need for a translator in
court proceedings and child's ethnicity. Similar to the situation with children needing
translators in court proceedings, African American and White mothers generally had little
need for translators in court proceedings, with only one mother per group needing
assistance in another language. However, 18.4% of Latina mothers (26 of 141 cases),
54.8% of Asian American/Pacific Islanders (17 of 31 cases), and 13.3% of Others (6 of
45 cases) needed translators in court proceedings. More striking than the child translator
(in court proceedings) results, many non-Black minority mothers need services and

63

assistance in their native language, especially Asian American/Pacific Islander mothers.
Please see Table 27.
Table 27: Mother's Need for Translator in Court Proceedings
Total

Black

White

Latino

Asian/PI

Other

Indicated

51
(12.7%)

1
(2.0%)

(0.7%)

I

26
(18.4%)

17
(54.8%)

6
(13.3%)

Not Indicated

352
(87.3%)

49
(98.0%)

135
(99.3%)

115
(81.6%)

14
(45.2%)

39
(86.7%)

Mother's Place of Birth
There was a significant relationship between mother's place of birth and child's
ethnicity. Again, consistent with the migration-related characteristics, African Americans
and Whites had low proportions of mothers born outside of the United States, and Latinas
and Asian American/Pacific Islanders had high proportion of mothers who were
immigrants. A majority of African American mothers (98%; 49 cases of 50) were born in
the U.S. as were 95.6% of White mothers (130 of 136 cases), nearly 75% of Latinas (105
of 141 cases), and 86.7% of Others (39 of 45 cases). Similar to the high proportion of
Asian American/Pacific Islander mothers needing translators in court proceedings,
approximately 70% of mothers from that ethnic group are immigrants. Please see Table
28.
Table 28: Mother's Place ofBirth
Total

Black

White

Latino

Asian/PI

Other

Indicated
Outside U.S.

71
(17.6%)

1
(2.0%)

6
(4.4%)

36
(25.5%)

22
(71.0%)

6
(13.3%)

Indicated U.S.
or Assumed
U.S.

332
(82.4%)

49
(98.0%)

130
(95.6%)

105
(74.5%)

9
(29.0%)

39
(86.7%)

Mother's Education Level
There was a marginally significant relationship [p = .061] between mother's
education and child's ethnicity A high proportion of mothers overall did not complete
high school (48.3%, or 116 of240 cases with valid information). Among groups, 40.5%
of African American mothers (15 of 37 cases), 38.6% of White mothers (27 of 70 cases),
57.8% of Latinas (52 of90 cases), 64.7% of Asian American/Pacific Islanders (11 of 17

64

cases), and 42.3% of Other mothers (11 of26 cases) had less than a high school level
education. Please see Table 29.
Table 29: Mother's Education Levela

Less Than
High School
Level
High School
Level or
Above
a

Total

Black

White

Latino

Asian/PI

Other

116
(48.3%)

15
(40.5%)

27
(38.6%)

52
(57.8%)

11
(64.7%)

11
(42.3%)

124
(51.7%)

22
(59.5%)

43
(61.4%)

38
(42.2%)

6
(35.3%)

15
(57.7%)

Based on 240 cases with valid information

Mother's Marital Status
There was a marginally significant relationship [p = .062] between mother's
marital status and child's ethnicity. Of note, many African American mothers were
single/never married (42.2%), many White mothers separated/divorced/widowed
(41.6%), and many Asian American/Pacific Islanders currently married (58.3%). Please
see Table 30.
Table 30: Mother's Marital Status8
Total

Black

White

Latino

Asian/PI

Other

Currently
Married

117
(34.9%)

(28.9%)

13

37
(32.7%)

37
(32.5%)

14
(58.3%)

16
(41.0%)

Single/ Never
Married

94
(28.1%)

19
(42.2%)

27
(23.9%)

33
(28.9%)

5
(20.8%)

10
(25.6%)

Separated/
Divorced/
Widowed

113
(33.7%)

10
(22.2%)

47
(41.6%)

38
(33.3%)

5
(20.8%)

13
(33.3%)

Other Status

11
(3.3%)

3
(6.7%)

2
(1.8%)

6
(5.3%)

0
(0.0%)

0
(0.0%)

a Based on 335 cases with valid information

65

Mother's Welfare Eligibility
There was a significant relationship·between a mother's welfare eligibility and
child's ethnicity. 34% of African American mothers (17 of 50 cases) are eligible, 14% of
Whites (19 of 136 cases), 23.4% of Latinas (33 of 141 cases), 19.4% of Asian
American/Pacific Islanders (6 of 31 cases), and 15 .6% of Others (7 of 45 cases). Please
see Table 31.
Table 31: Mother's Welfare Eligibility
Total

Black

White

Latino

Asian/Pl

Other

Indicated

82
(20.3%)

17
(34.0%)

19
(14.0%)

33
(23.4%)

6
(19.4%)

7
(15.6%)

Not Indicated

321
(79.7%)

33
(66.0%)

117
(86.0%)

108
(76.6%)

25
(80.6%)

38
(84.4%)

Maternal Substance Abuse
There was a significant relationship between maternal substance abuse and child's
ethnicity. A high proportion of mothers overall were substance abusers (54.6%, or 220 of
403 cases). Among groups, 62.0% of African American mothers (31 of50 cases) were
indicated as having problems with substance abuse. 55.9% of Whites (76 of 136 cases),
55.3% of Latinas (78 of 141 cases), 29.0% of Asian American/Pacific Islanders (9 of 31
cases), and 57.8% of Other mothers (26 of 45 cases) were suffering from substance
abuse. Please see Table 32.
Table 32: Maternal Substance Abuse
Total

Black

White

Latino

Asian/PI

Other

Indicated·

220
(54.6%)

31
(62.0%)

76
(55.9%)

78
(55.3%)

9
(29.0%)

26
(57.8%)

Not Indicated

183
(45.4%)

19
(38.0%)

60
(44.1%)

63
(44.7%)

22
(71.0%)

19
(42.2%)

Maternal Mental Health Problems
There was a significant relationship between identified maternal mental health
problems and child's ethnicity. Among groups, identified mental health problems were
indicated among 14.0% of African American mothers (7 of 50 cases), 25.0% of Whites

66

(34 of 136 cases), 12.8% of Latinas (18 of 141 cases), 25.8% of Asian American/Pacific
Islanders (8 of31 cases), and 13.3% of Other mothers (6 of 45 cases). Please see Table
33.
Table 33: Maternal Mental Health Problems
Total

Black

White

Latino

Asian/PI

Other

Indicated

73
(18.1%)

7
(14.0%)

34
(25.0%)

18
(12.8%)

8
(25.8%)

6
(13.3%)

Not Indicated

330
(81.9%)

43
(86.0%)

102
(75.0%)

123
(87.2%)

23
(74.2%)

39
(86.7%)

Mother's Participation in Mental Health Service System
A significant relationship was found between mother's participation in mental
health services and child's ethnicity. Among groups, 4.0% of African American mothers
(2 of 50 cases) participated in the mental health service system, as had 15.4% of Whites
(21 of 136 cases), 7.8% of Latinas (11 of 141 cases), 19.4% of Asian American/Pacific
Islanders (6 of 31 cases), and 8.9% of Other mothers (4 of 45 cases). Please see Table 34.
Table 34: Mother's Participation in Mental Health Service System
Total

Black

White

Latino

Asian/PI

Other

Indicated

44
(10.9%)

2
(4.0%)

21
(15.4%)

11
(7.8%)

6
(19.4%)

4
(8.9%)

Not Indicated

359
(89.1%)

48
(96.0%)

115
(84.6%)

130
(92.2%)

25
(80.6%)

41
(91.1%)

Maternal Criminal History
There was a significant relationship between maternal criminal history and child's
ethnicity. A history of maternal criminal behavior marked a majority of the cases (53.8%,
or 217 of 403 cases). Among groups, a criminal history was indicated in 72.0% of
African American mothers (36 of 50 cases), 48.5% of Whites (66 of 136 cases), 56.0% of
Latinas (79 of 141 cases), 25.8% of Asian American/Pacific Islanders (8 of 31 cases), and
62.2% of Other mothers (28 of 45 cases). Please see Table 35.

67

Table 35: Maternal Criminal History
Total

Black

White

Latino

Asian/PI

Other

Indicated

217
(53.8%)

36
(72.0%)

66
(48.5%)

79
(56.0%)

8
(25.8%)

28
(62.2%)

Not Indicated

186
(46.2%)

14
(28.0%)

70
(51.5%)

62
(44.0%)

23
(74.2%)

17
(37.8%)

Number of Maternal Criminal Episodes
Corroborating the results of maternal criminal history, incarceration, and
involvement with the criminal justice system, a significant relationship was found
between the number of the mother's criminal episodes and child's ethnicity. Among
groups, African American mothers had an average of 3.46 (sd = 3.99) criminal episodes,
Whites 1.71 (sd = 2.62), Latinas 2.90 (sd = 5.12), Asian American/Pacific Islanders_0.94
(sd = 2.17), and Others 2.16 (sd = 3.46). Please see Table 36.
Table 36: Number of Maternal Criminal Episodes3
Number of
Maternal
Criminal
Episodes
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

2.34
(3.95)

3.46
(3.99)

1.71
(2.62)

2.90
(5.12)

0.94
(2.17)

2.16
(3.46)

a Based on 403 cases with valid infonnation

Maternal Incarceration
Similar to maternal criminal history, a significant relationship was found between
maternal incarceration and child's ethnicity. Among groups, 20.0% of African American
mothers (10 of 50 cases) were incarcerated at some point before the incident of child
abuse or neglect. 7.4% of Whites (10 of 136 cases), 10.6% of Latinas (15 of 141 cases),
6.5% of Asian American/Pacific Islanders (2 of31 cases), and 2.2% of Other mothers (1
of 45 cases) had been incarcerated at some point in their past. Please see Table 37.

68

Table 37: Maternal Incarceration
Total

Black

White

Latino

Asian/Pl

Other

Indicated

38
(9.4%)

(20.0%)

10

10
(7.4%)

15
(10.6%)

2
(6.5%)

1
(2.2%)

Not Indicated

365
(90.6%)

40
(80.0%)

126
(92.6%)

126
(89.4%)

29
(93.5%)

44
(97.8%)

Mother's Participation in Criminal Justice System
Corroborating the results of maternal criminal history and incarceration, a
significant relationship was found between mother's participation in the criminaljustice
system and child's ethnicity. Among groups, 38.0% of African American mothers (19 of
50 cases) had participated in the criminal justice system. 16.2% of Whites (22 of 136
cases), 32.6% of Latinas (46 of 141 cases), 16.l % of Asian American/Pacific Islanders (5
of 31 cases), and 28.9% of Other mothers (13 of 45 cases) had been in the criminal
justice system. Please see Table 38.
Table 38: Mother's Participation in Criminal Justice System
Total

Black

White

Latino

Asian/PI

Other

Indicated

105
(26.1%)

19
(38.0%)

22
(16.2%)

46
(32.6%)

(16.1%)

5

13
(28.9%)

Not Indicated

298
(73.9_%)

31
(62.0%)

114
(83.8%)

95
(67.4%)

26
(83.9%)

32
(71. 1%)

Mother's Current Living Situation
Similar to the results of maternal-marital status, a significant relationship was
found between mother's current living situation and child's ethnicity. Among groups,
very few African American mothers (4.0%) were living with a spouse (2 of 50 cases). By
contrast, 24.3% of Whites (33 of 136 cases), 18.4% ofLatinas (26 of 141 cases), 32.3%
of Asian American/Pacific Islanders (10 of 31 cases), and 28.9% of Other mothers (13 of
45 cases) were living with their spouse. Please see Table 19. Again of note,. Please see
Table 39.

69

Table 39: Mother's Current Living Situation
Total

Black

White

Latino

Asian/Pl

Other

With Spouse

84
(20.8%)

2
(4.0%)

33
(24.3%)

26
(18.4%)

IO

(32.3%)

13
(28.9%)

Not With
Spouse /Not
Indicated

319
(79.2%)

48
(96.0%)

103
(75.7%)

115
(81.6%)

21
(67.7%)

32
(71.1%)

Number of Children Living with Mother
There was a significant difference in the number of children living with the
mother across child's ethnic group. Latina mothers had an average of 2. 75 children living
with them (sd = 1.49), Other mothers 2.50 children (sd = 1.15), Asian American/Pacific
Islanders 2.33 children (sd = 1.20), African Americans 2.29 children (sd = 1.30) and
Whites 2.13 children (sd = 1.18). Please see Table 40.
Table 40: Number of Children Living with Mother8
Number of
Children
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

2.42
(1.34)

2.29
(1.30)

2.13
(1.18)

2.75
(1.49)

2.33
(1.20)

2.50
(1.15)

a Based on 245 cases with valid information

General Family-Related Factors ( Domestic Violence, Income Problems, and
Generational Involvement in the CWS)
There was not a significant statistical relationship between family factors
including domestic violence, income problems and generational involvement in the CWS
and ethnic group membership. However, it is notable that domestic violence was
indicated in almost half (45.4%) of the cases. About one-fifth (21.3%) of families
sampled indicated income problems and 7.7% of the 403 cases were families with
generational involvement in the CWS. Please see Table 41.

70

Table 41: Family Factors
Total
Domestic Violence
Income Problems
Generational
Involvement in the

cws

Freouency and% of Sub ,roup
White
Black
Latino Asian/PI

62
183
20
69
(45.4%) (40.0%) (50.7%) (44.0%)
28
12
27
86
(21.3%) (28.0%) (20.6%) (19.1%)
0
11
14
31
(7.7%) (0.0%) (8.1%) (9.9%)

12
(38.7%)
8
(25.8%)
1
(3.2%)

Other
20
(44.4%)
9
(20.0%)
5
(11.1%)

Perpetrator's Relationship to the Child

There was a significant relationship between the perpetrator's relationship to the
child and child's ethnicity. Of note, many African American CWS cases involved the
mother only as the perpetrator of child abuse or neglect (59.6%, or 28 of 47 cases). In
comparison, many of the White, Hispanic/Latino, Asian American/Pacific Islander, and
Other cases involved a combination of perpetrators (mothers and others, including
fathers): White (48.5%), Hispanic/Latino (49.6%), Asian American/Pacific Islander
(50.0%), and Other (60.5%). Also, a notable proportion of Asian American/Pacific
Islander cases involved the father as the lone perpetrator. Please see Table 42.

71

Table 42: Perpetrator's Relationship to the Childa
Total

Black

White

Latino

Asian/PI

Other

Mother Only
(including a
Step-Mother)

144
(36.6%)

28
(59.6%)

42
(31.8%)

52
(36.9%)

9
(30.0%)

13
(30.2%)

Mother and
Others
(including the
Father)

191
(48.6%)

16
(34.0%)

64
(48.5%)

70
(49.6%)

15
(50.0%)

26
(60.5%)

Father Only
(including a
Step-Father)

31
(7.9%)

1
(2.1%)

14
(10.6%)

7
(5.0%)

6
(20.0%)

3
(7.0%)

Other
Perpetrator

27
(6.9%)

2
(4.3%)

12
(9.1%)

12
(8.5%)

0
(0.0%)

1
(2.3%)

a Based on 393 cases with valid information

Children and Number of Siblings
There was a significant difference in the number of siblings across child's ethnic group.
African American children had an average of 2. 79 siblings (sd = 2.07), Others 2. 79 siblings (sd =
1.89), Hispanics/Latinos 2.64 siblings (sd = 1.85), Whites L98 siblings (sd = 1.63), and
Asians/Pis 1.77 siblings (sd = 1.12). Please see Table 43.
Table 43: Number of Siblingsa
Number of
Siblings
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

2.39
(1.80)

2.79
(2.07)

1.98
(1.63)

2.64
(1.85)

1.77
(1.12)

2.79
(1.89)

a Based on 368 cases with valid information

72

Children and Number of Siblings in the Child Welfare System
There was a significant difference in the number of siblings in the CWS across child's
ethnic group. African American children had an average of 1.81 siblings in the CWS (sd = 1.75),
Others 1.76 siblings in CWS (sd = 1.73), Hispanics/Latinos 1.60 siblings in CWS (sd = 1.59),
Whites 1.13 siblings in CWS (sd = 1.27), and Asians/Pis 1.07 siblings in CWS (sd = 0.92).
Please see Table 44.
Table 44: Number of Siblings in CWS 8
Number of
Siblings in the

cws

Avg.
(sd)

a Based

Total

Black

White

Latino

Asian/PI

Other

1.45
(1.51)

1.81
(1.75)

1.13
(1.27)

1.60
(1.59)

1.07
(0.92)

1.76
(1.73)

on 368 cases with valid information

73

2.

System-Related Characteristics by Ethnic Group

The majority of the data in this section are derived from the CWS/CMS database of
closed cases. Longitudinal variables were constructed to depict the child's history in the CWS.

Number of Prior Referrals
There was not a statistically significant relationship between number of prior referrals
and ethnicity. An average of2.46 prior referrals (sd = 3.14) were made before the case was
opened for the current episode. Please see Table 45.
Table 45. Number of Prior Referrals By Ethnic Group
Prior
Referralsa
Avg.
(sd)
a

Total

Black

White

Latino

Asian/PI

Other

2.46
(3.14)

2.14
(2.71)

2.64
(3.27)

2.46
(3.25)

1.62
(2.14)

2.81
(3.37)

Based on 355 cases with valid information

Reporter of the Child Abuse or Neglect
There was a significant relationship between who reported the abuse/neglect incident and
ethnicity. Law enforcement officers and legal professionals were the most common reporters
within all groups except Others, and were especially common reporters for Whites and African
Americans. For African Americans, personnel at medical sites were common reporters.
Reporters from school sites were more common for Hispanics/Latinos and Asians than other
groups. For Others, a category including neighbors, friends, and various other reporters was most
common (38.9% of the cases). Please see Table 46.

74

Table 46: Reporter of Child Abuse or Neglecta
Total

Black

White

Latino

Asian/PI

Other

63
(19.3%)

7
(17.1%)

22
(20.4%)

24
(20.2%)

5
(21.7%)

(13.9%)

Law
Enforcement /
Legal
Professional

123
(37.6%)

17
(41.5%)

47
(43.5%)

41
(34.5%)

8
(34.8%)

(27.8%)

Personnel at
Medical Site

37
(11.3%)

9
(22.0%)

10
(9.3%)

(10.9%)

13

2
(8.7%)

3
(8.3%)

Personnel at
School Site

32
(9.8%)

1
(2.4%)

7
(6.5%)

18
(15.1%)

3
(13.0%)

3
(8.3%)

Neighbor/
Friend/
Citizen /CPS/
Agency/Day
Care/fherapist
Anonymous I
Other

70
(21.4%)

7
(17.1%)

22
(20.4%)

23
(19.3%)

4
(17.4%)

14
(38.9%)

Multiple
Reporters

2
(0.6%)

0
(0.0%)

0
(0.0%)

0
(0.0%)

1
(4.3%)

1
(2.8%)

Relative/
Family
Member / Self

a

5

10

Based on 327 cases with valid information

Reason for Removal

There was not a significant relationship between the initial reason for removal and
ethnicity. In the sample, "Other" types of maltreatment were the most common reason for
removal amongst all groups. Asian/Pacific Americans (35.5%, 11 out of 31) and
Hispanics/Latinos (24.4%, 33 out of 135) had higher rates than other groups of physical abuse.
Please see Table 47.

75

Table 47: Reason for Removal8
Total

Black

White

Latino

Asian/Pl

Other

General
Neglect

50
(12.9%)

7
(14.3%)

15
(11.7%)

22
(16.3%)

2
(6.5%)

4
(9.1%)

Severe
Neglect

43
(11.1%)

7
(14.3%)

11

(8.6%)

14
(10.4%)

4
(12.9%)

7
(15.9%)

Physical
Abuse

86
(22.2%)

9
(18.4%)

24
(18.8%)

33
(24.4%)

(35.5%)

11

9
(20.5%)

Sexual Abuse

30
(7.8%)

3
(6.1 %)

13
(10.2%)

(8.1%)

11

2
(6.5%)

1
(2.3%)

Emotional
Abuse

20
(5.2%)

2
(4.1%)

7
(5.5%)

4
(3.0%)

2
(6.5%)

(11.4%)

Other
Maltreatment

158
(40.8%)

21
(42.9%)

58
(45.3%)

51
(37.8%)

10
(32.2%)

18
(40.9%)

a Based

5

on 3 87 cases with valid information

Case Assignment after Jurisdictional/Dispositional Hearing
There was not a significant relationship between case assignment after the
jurisdictional/dispositional hearing and ethnicity. Overall, most cases were assigned to family
reunification (71.1%, 278 of391 cases). African American children had the highest percentage
of family reunification (80.9%, 38 of 47 cases), while Asian American/Pacific Islander children
had the highest percentage of family maintenance (29.0%, 9 of 3 I cases). Twenty-five cases
were assigned neither to family reunification nor to family maintenance but to other disposition
categories. The largest numbers of cases assigned neither to family reunification nor family
maintenance ( 8) represented cases for which services were denied. Please see Table 48.

76

Table 48: Case Assignment after Jurisdictional/Dispositional Hearing8
Family
Maintainence

Total
88
(22.5%)

Black
7
(14.9%)

White
34
(25.8%)

Latino
27
(19.4%)

Asian/PI
9
(29.0%)

Other
11
(26.2%)

Family
Reunification

278
(71.1%)

38
(80.9%)

91
(68.9%)

101
(72.7%)

19
(61.3%)

29
(69.0%)

Other

25
(6.4%)

2
(4.3%)

7
(5.3%)

11
(7.9%)

3
(9.7%)

2
(4.8%)

8

Based on 391 cases with valid information

Additional Analysis:
Too few people had been assigned to categories other than family reunification and
family maintenance to permit further analysis of cases falling into those groups. In what follows,
we examine racial and ethnic differences in the 366 cases assigned either to family maintenance
or family reunification.
Through statistical analysis, we sought to understand racial and ethnic differences in
assignment to family maintenance vs. family reunification. Our approach was not only to
consider differences in race and ethnicity themselves as we had in our previous analyses, but also
to consider a number of factors which we had come to understand were associated with race and
ethnicity and which might explain any racial and ethnic difference we might document The
question for this analysis, then, was whether underlying differences in child and family status
revealed earlier in our work accounted for racial and ethnic differences in assignment.
To achieve this purpose we used Logistic Regression. This is a procedure that permits
assessment of racial and ethnic differences in assignment to family maintenance vs. family
reunification. It does so both before and after controlling for characteristics shown to be
important apart from race and ethnicity. It addresses the question: If racial and ethnic differences
appear, then what other characteristics associated with children, parents, and families of might
explain them?
We identified a list of characteristics already shown to be linked to which racial and
ethnic group to which a child belonged. These are: Whether the child was born in the United
States, whether the child had suffered sexual abuse, whether emotional problems were indicated
for the child, his or her age, the number of prior referrals, whether the mother needed a translator

77

in court proceedings, mothers' criminal history, and whether the case had been reported by
relatives/family members/ self or by law enforcement.
The analysis proceeded in steps. Racial and ethnic differences were ~onsidered first. This
was done to establish whether individual racial and ethnic groups differed from Whites in their
chances being assigned to family maintenance.
Other characteristics listed above entered in later steps. This was done to determine
which of the characteristics might explain racial and ethnic differences identified in the first step.
Results were as follows. African Americans were found to differ from Whites at a
statistically significant level (B=l .22, p<.03). This result indicated that African Americans were
more likely than whites, at a statistically significant level, to be assigned to Family
Reunification. ~o other group was found to be statistically significantly different from Whites.
The difference between African Americans and white persisted after controlling for all of
the characteristics noted above. Several of the characteristics were indeed significantly
associated with assignment to family reunification. These included reporting by relatives/family
members/self(B=l.59, p<.05), older age of children (B=.10, p<.05), and jointly reporting by law
enforcement and mother's criminal history (B=.246, p<.05). However, African Americans
continued to be more likely to be assigned to Family Reunification after adjustment for the
impact of these factors.

Initial Out of Home Placement Facility Type
There was not a significant relationship between initial out-of-home placement facility
type and ethnicity. In the sample, a relative home was the most frequently occurring placement
type for all groups except Asian American/Pacific Islander children who were highest in foster
family home placement (32.3%, 10 of31 cases). African American children also were relatively
high in foster family home placement (22.4%, 11 out of 49). Please see Table 49.

78

T a ble 49 Im·r1a10 Ut 0 f Home PlacementFac11ty
Tt Typea
Total
Black
White

Latino

Asian/PI

Other

Foster Family
Home

72
(18.6%)

11
(22.4%)

20
(15.6%)

18
(13.3%)

10
(32.3%)

13
(29.5%)

Group Home

34
(8.8%)

4
(8.2%)

15
(11.7%)

IO
(7.4%)

2
(6.5%)

3
(6.8%)

72
(18.6%)

7
(14.3%)

22
(17.2%)

27
(20.0%)

7
(22.6%)

9
(20.5%)

Relative Home

149
(38.5%)

21
(42.9%)

49
(38.3%)

58
(43.0%)

7
(22.6%)

14
(31.8%)

Foster Family
Agency

40
(10.3%)

5
(10.2%)

17
(13.3%)

13
(9.6%)

4
(12.9%)

1
(2.3%)

Guardian
Home/ Court
Specified
Home

20
(5.2%)

1
(2.0%)

5
(3.9%)

9
(6.7%)

1
(3.2%)

4
(9.1%)

Children's
Shelter/
Receiving
Home/NonEA/AFDC

a Based on 387 cases with valid information

Number of Times Removed from Family in Current Episode
There was a significant relationship between the number of times removed from family in
current episode and ethnicity. Across all ethnic groups, children were removed an average of
1.45 times (sd = 0.74). African American children had the highest average number of times
removed from family in current episode with 1.67 times (sd = 0.90), and Asian American/Pacific
Islander children had the lowest average with 1.26 times (sd = 0.51). Please see Table 50.
Table 50: Number of Times Removed from Family in Current Episodea

Average
(sd)
0

Total

Black

White

Latino

Asian/PI

Other

1.45
(0.74)

1.67
(0.90)

1.36
(0.68)

1.44
(0.68)

1.26
(0.51)

1.64
(0.92)

Based on 388 cases with valid information

79

Number of Unique Placement Homes in Current Episode
There was not a significant relationship between number of unique placement homes in
current episode and ethnicity. Across all ethnic groups, there was an average of 4.54 (sd = 4.67)
unique placement homes in the child's current episode. African American children had the
highest average of unique placement homes in current episode with 5.35 unique placement (sd =
6.48), and Asian American/Pacific Islander children had the lowest average with 3 .29 unique
placements (sd = 3.22). Please see Table 51.

Table 51: Number of Unique Placement Homes in Current Episodea

Average
(sd)
a Based

Total

Black

White

Latino

Asian/PI

Other

4.54
(4.67)

5.35
(6.48)

4.70
(4.60)

4.44
(4.26)

3.29
(3.22)

4.36
(4.59)

on 388 cases with valid information

Average Stay (in days) per Placement Facility in Current Episode
There was a significant relationship between the average stay (in days) per placement
facility in current episode and ethnicity. For all children in the sample, the average number of
days a child stayed in one placement in their current episode was 370.77 (sd = 710.59). African
American children had the lengthiest stay per placement with an average of 678.94 days (sd =
1121.34), and Other children had the shortest stay with and average of221.67 days (sd =
278.71). Please see Table 52.
Table 52: Average stay (in days) per Placement Facility in Current Episodea

Average
(sd)
8

Total

Black

White

Latino

Asian/PI

Other

370.77
(710.59)

678.94
(1121.34)

222.54
(410.37)

394.57
(671.36)

. 608.49
(1146.98)

221.67
(278.71)

Based on 388 cases with valid information

80

Number of Placements in Current Episode
There was not a significant relationship between the number of placements in current
episode and ethnicity. Across ethnic groups the average number of placements in current episode
was 3.31 (sd = 3.34). White children had the highest average number of placements with 3.77 (sd
= 4.06), and Hispanic/Latino children had the lowest average with 2.99 (sd = 2.78) placements in
·
current episode. Please see Table 53.
Table 53: Number of Placements in Current Episodea

Average
(sd)
a

Total

Black

White

Latino

Asian/PI

Other

3.31
(3.34)

3.27
(3.23)

3.77
(4.06)

2.99
(2.78)

3.00
(3.16)

3.20
(2.83)

Based on 387 cases with valid information

Total length of time (in months) in Out of Home Placement in Current Episode
There was a significant relationship between the total length of time. (in months) in outof-home placement (for all placement types) in current episode and ethnicity. Across ethnic
groups, children spent an average of 18.01 months (sd = 25.71) in out-of-home-placement. Asian
American/Pacific Islander children spent the lengthiest time in out-of-home placement with an
average of29.65 months (sd = 40.51), and White children spent the shortest amount of time in
out-of-home placement with an average of 12.50 months (sd = 15.85). Please see Table 54.

Table 54: Total Length ofTime (in months) in Out Of Home Placement in Current Episode8

Average
(sd)
a Based

Total

Black

White

Latino

Asian/PI

Other

18.01
(25.71)

24.88
(34.71)

12.50
(15.85)

19.49
(25.89)

29.65
(40.51)

13.58
(19.49)

on 385 cases with valid information

Number of Episodes
There was not a significant relationship between the number of episodes and ethnicity.
On average, ch~ldren across ethnic groups had 1.28 episodes (sd = 0.54). Hispanic/Latino

81

children had the highest average number of episodes, with 1.32 (sd = 0.54), and White children
had the lowest average number of episodes with 1.20 (sd = 0.49). Please see Table 55.
Table 55: Number ofEpisodes8

Average
(sd)
a Based

Total

Black

White

Latino

Asian/PI

Other

1.28
(0.54)

1.26
(0.49)

1.20
(0.42)

1.32
(0.64)

1.32
(0.54)

1.36
(0.57)

on 392 cases with valid information

Length of Case in Years
There was a· significant relationship between the length of the case in years and ethnicity.
The sample had an average case length of 3.30 years (sd = 3.47). African American children had
the highest average case length with 5.04 years (sd = 4.64) and Other children had the lowest
average case length with 2.55 years (2.80). Please see Table 56.
Table 56: Length of Case in Years8

Average
(sd)
a

Total

Black

White

Latino

Asian/Pl

Other

3.30
(3.47)

5.04
(4.64)

2.60
(2.75)

3.41
(3.52)

4.06
(3.69)

2.55
(2.80)

Based on 392 cases with valid information

Total Nµ.mber of Workers Assigned (over time) in the Current Episode
There was not a significant relationship between the total number of workers assigned
(over time) in the current episode and ethnicity. Across all ethnic groups, children had an
average of 4.74 workers (sd = 2.03) assigned to their case over time. Hispanic/Latino children
had the most workers with 4.88 (sd = 2.08), and Asian American/Pacific Islander children had
the fewest number of workers with 4.42 (sd = 1.61). Please see Table 57.
Table 57: Total Number of Workers Assigned (over time) in the Current Episode2

Average
(sd)
11

Total

Black

White

Latino

Asian/PI

Other

4.74
(2.03)

4.58
(2.23)

4.83
(2.06)

4.88
(2.08)

4.42
(1.61)

4.47
(1.82)

Based on 396 cases with valid information

82

Last Out-of-Home Placement Type

There was a significant relationship between last placement type and ethnicity.
Asian/Pacific (37.5%, 9 of24 cases) and African American children (26.2%, 11 out of 42) were
more likely than children from other groups to be placed in a foster family home. Asian/Pacific
children were more also likely to be placed at a foster family agency (20.8%, 5 out of24).
Hispanic/Latino children were especially likely to be placed at a relative home (43.8%, 53 of 121
cases), or a children's shelter (18.2%, 22 out of 121 cases. Please see Table 58.
Table 58: Last Out-of-Home Placement Type8
Total

Black

White

Latino

Asian/PI

Other

Foster Family
Home

66
(19.1 %)

11
(26.2%)

20
(16.5%)

13
(10.7%)

9
(37.5%)

13
(35.1%)

Group Home

33
(9.6%)

5
(11.9%)

12
(9.9%)

(9.1%)

11

2
(8.2%)

3
(8.1%)

Children's
Shelter

50
(14.5%)

2
(4.8%)

17
(14.0%)

22
(18.2%)

3
(12.5%)

6
(16.2%)

Relative
Home

131
(38.0%)

16
(38.1%)

48
(39.7%)

53
(43.8%)

3
(12.5%)

(29.7%)

Foster Family
Agency

43
(12.5%)

6
(14.3%)

18
(14.9%)

13
(10.7%)

5
(20.8%)

1
(2.7%)

Guardian
Home/ Court
Specified
Home

22
(6.4%)

2
(4.8%)

9
(7.4%)

9
(7.4%)

2
(8.3%)

3
(8.1 %)

11

a Based on 345 cases with valid information

Age at Time of Case Closure in Current Episode

There was not a significant relationship between the child's age at time of case closure in
current episode and ethnicity. The average age at time of closure for the sample was 10.55 yearsof-age (6.35). Hispanic/Latino children were the oldest at the time of case closure (11.39, sd =

83

6.21), and African American children were the youngest at time of case closure (9.36). Please
see Table 59.
Table 59: Age at Time of Case Closure in Current Episodea

Average
(sd)
a Based

Total

Black

White

Latino

Asian/PI

Other

10.55
(6.35)

9.36
(6.36)

10.23
(6.27)

11.39
(6.21)

11.36
(6.34)

9.61
(6.84)

on 3 84 cases with valid infonnation

Service Type at Closure of Case
There was a significant relationship between the service type at closure of case and
ethnicity. White children had the highest percentage of family maintenance services (52.2%, 71
of 136 cases), as well as the highest percentage of family reunification (12.5%, 17 of 136 cases).
African American (52.0%, 26 of 50 cases), Hispanic/Latino (50.4%, 71 out of 141), and Asian
American/Pacific Islander (48.4%, 15 out of 31) children had the highest percentage of
permanent placement services at closure of their case. Please see Table 60.
Table 60: Service Type at Case Closurea
Total

Black

White

Latino

Asian/PI

Other

Emergency
Response

9
(2.2%)

1
(2.0%)

2
(1.5%)

3
(2.1%)

0
(0.0%)

3
(6.7%)

·Family
Maintenance

186
(46.2%)

23
(46/0%)

71
(52.2%)

56
(39.7%)

14
(45.2%)

22
(48.9%)

Family
Reunification

34
(8.4%)

0
(0.0%)

17
(12.5%)

11
(7.8%)

2
(6.5%)

4
(8.9%)

Permanent
Placement

174
(43.2%)

26
(52.0%)

46
(33.8%)

71
(50.4%)

15
(48.4%)

16
(35.6%)

a Based

on total sample of 403 cases

84

Case Closure Type
There was not a significant relationship between case closure type and ethnicity. In the
sample, Asian American/Pacific Islander children had the highest percentage for court ordered
termination and other case closure (45.2%, 14 of31 cases) and emancipation (22.6%, 7 of 31
cases). White children (30.5%, 40 of 131 cases) had the highest percentage for family stabilized.
Hispanic/Latino children had the highest percentage of guardianship or placement with relative
(15.9%, 21 of 132 cases). Other children had the highest percentage for reunification with parent
or guardian (11.4%, 5 of 44 cases). Please see Table 61

85

Table 61: Case Ciosure Type8
Total

Black

White

Latino

Asian/PI

Other

Adoption

16
(4.1%)

4
(8.3%)

6
(4.6%)

2
(1.5%)

2
(6.5%)

2
(4.5%)

Emancipation

57
(14.8%)

6
(12.5%)

11
(8.4%)

25
(18.9%)

7
(22.6%)

8
(18.2%)

Family
Stabilized
(FM)

98
(25.4%)

13
(27.1%)

40
(30.5%)

31
(23.5%)

5
(16.1%)

9
(20.5%)

44
(11.4%)

5
(10.4%)

13
(9.9%)

21
(15.9%)

2
(6.5%)

3
(6.8%)

25
(6.5%)

3
(6.3%)

11
(8.4%)

6
(4.5%)

0
(0.0%)

(11.4%)

27
(7.0%)

6
(12.5%)

7
(5.3%)

10
(7.6%)

1
(3.2%)

3
(6.8%)

119
(30.8%)

11
(22.9%)

43
(32.8%)

37
(28.0%)

14
(45.2%)

14
(31.8%)

Guardianship
established or
placement
with relative
Reunified with
parent or
guardian,
court or noncourt specified
Incarceration,
runaway or
medical
services
Court ordered
termination
and other

a Based on 386 cases with valid information

5

86

3.

Services and Other Interventions Ordered at the
Jurisdictional/Dispositional Hearing by Ethnic Group

The data in this section are derived from the case record review database.
We sought to document the frequency of occurrence of services and other interventions
ordered for children, mothers, and fathers at the Jurisdictional/Dispositional Hearing and to
determine the extent of racial and ethnic differences in services and interventions. From our case

record review we discovered that many services and other interventions were noted only rarely
as having been ordered. These were:
Child:

Not allowed visits (n=l; .2%)
Kin visits supervised or unsupervised, ordered (n=21; 5.2%)
Not allowed kin visits (n=l; 0.2%)
Random drug/alcohol testing (n=0; 0%)
Substance abuse/alcohol treatment ordered (n=2; .5%)
Orders not to possess alcohol or drugs: ordered (n=l; .2%)
12-step (n=l; .2%)
Support group ordered (n=5; 1.2%)
Mental health system ordered (n=3; .7%)
Health system ordered (n=19; 4.7%)
Disability-related service ordered (n=0; 0%)
Anger management ordered (n=0; 0%).
DV for victim ordered (N=0; 0%).
No criminal activity ordered (n=0; 0%)
Insure school attendance progress ordered (n=4; 1.0%)
Family conference ordered (n=4; 1.0%).
Other service ordered (n=l 3; 3.2%).
Mother:
Unsupervised visits ordered (n=l4; 3.5%)
Contingent visits ordered (n=4; 1.0%)
Not allowed visits (n=3;0.7%).
Kin visits ordered (n=3; 0.7%)
Kin visits disallowed (n=0; 0%)
Orders not to use or possess drugs or alcohol (n=9; 2.2%)
Support group ordered (n=13; 3.2%)
Mental health system ordered (n=5; 1.2%).
Health system ordered (n=lO; 2.5%)
Disability-related service ordered (n=0; 0%)

87

Anger management ordered (n=9; 2.2%).
Cooperate with officials ordered (n=l 8; 4.5%).
No physical punishment ordered (n=7; 1.7%)
No criminal activity ordered (n=6; 1.5%).
Improve economic employment status ordered (n=3; 0.7%).
Improve housing situation ordered (n=22; 5.5%).
Insurance of school attendance and progress ordered (n=5; 1.2%).
Family conference ordered (n=5; 1.2%).
Home supervision ordered (n=IO; 2.5%).
Other service ordered (n=l3; 3.2%).

Father:
Unsupervised visits ordered (n=l5; 3.7%)
Contingent visits ordered (n=9; 2.2%)
Not allowed visits (n=2 l; 5.2%).
Kin visits ordered (n=0; 0%)
Kin visits disallowed (n=0; 0%)
·Orders not to use or possess drugs or alcohol (n=6; 1.5%)
Support group ordered ( n=7; I. 7%)
Mental health system ordered (n=5; 1.2%).
Health system ordered (n=3; 0.7%)
Psychological evaluation ordered (n=l 9; 4.7%).
Disability-related service ordered (n=0; 0%)
Anger management ordered (n=l4; 3.5%).
Cooperate with officials ordered (n=7; 1.7%).
No physical punishment ordered (n=4; 1.0%)
No criminal activity ordered (n=4; 1.0%).
Improve economic employment status ordered (n=l; 0.2%).
Improve housing situation ordered (n=l4; 3.5%).
Insurance of school attendance and progress ordered (n=0; 0%).
Family conference ordered (n=2; 0.2%).
Home supervision ordered (n=0; 0%).
Other service ordered (n=5; 1.2%).
Infrequently reported services magnify the importance of errors in reporting and provide
little opportunity for the racial and ethnic differences to appear. We focus on services most
commonly recorded, defined as indicated for I 0% or more of the sample.

Child: Individual Counseling
Racial and ethnic differences in individual counseling were statistically significant (ChiSquare=l 4.6, p<.01). Individual counseling was recorded as ordered for more than 46% of white
children (63 out of 136), and 40% of children classified as other (18 out of 45). It was recorded

88

for 35.5% of Asian American/Pacific Islander children (11 out of 31), about 32% of
Hispanic/Latino children (45 out of 141 ), and 18% of African American children (9 out of 50).
The statistically significant difference parallels differences in indicated rates of mental health
problems, and indicates that Asian American/Pacific Islander children were less likely, and
Hispanic/Latino and African American children far less likely than whites to be ordered into
individual counseling. Please see Table 62.
Table 62. Child: Individual Counseling
Total

Black

White

Latino

Asian/PI

Other

Ordered

146
(36.2%)

9
(18.0%)

63
(46.3%)

45
(31.9%)

11
(35.5%)

18
(40.0%)

Not
Indicated

257
(63.8%)

41
(82.0%)

73
(53.7%)

96
68.1%

20
(64.5%)

27
(60.0%)

Child: Other Counseling
Racial and ethnic differences in whether or not "other" counseling was ordered were
significant (Chi-Square=l 1.7, p<.05). Counseling classified as other- than individual was
recorded as ordered for more than 35% (11 out of 45) Asian American/Pacific Islander children
and 20% of white children (28 out of 136) and 18.4% of Hispanic/Latino children (26 out of
141 ). Other counseling was ordered for 6% of African American children (3 out of 50). Please
see Table 63.
Table 63. Child: Other Counseling
Total

Black

White

Latino

Asian/PI

Other

Ordered

79
(19.6%)

3
(6%)

28
(20.6%)

26
(18.4%)

11
(35.5%)

11
(24.4%)

Not
Indicated

324
(80.4%)

47
(94%)

108
(79.4%)

115
(81.6%)

20
(64.5%)

34
(75.6%)

Child: Psychological Evaluation

89

Racial and ethnic differences in rates of psychological evaluation were statistically
significant (Chi-Square=7.9, p<.09). Psychological evaluation was recorded as ordered for 14%
of white children (19 out of 136), and less than 8% of Hispanic/Latino children, 6.5% of Asian
American/Pacific Islander children (2 out of 31 ), and 2% of African American children (1 out of
50). Differences in psychological evaluation are like those found for individual counseling, and
may reflect the interrelationship between evaluating and treating psychological problems;
evaluation may either precede or follow individual counseling. Please see Table 64.
Table 64. Child: Psychological Evaluation

Ordered
Not
Indicated

Total

Black

White

Latino

Asian/PI

Other

39
(9.7%)

1
(2.~%)

19
(14.0%)

11
(7.8%)

2
(6.5%)

6
(13.3%)

364
(90.3%)

49
(98.0%)

117
(86.0%)

130
(92.2%)

29
(93.5%)

39
(86.7%)

Mother: Parent Education Ordered
Parent education was ordered for 283 mothers representing 70.2% of the sample. Racial
and ethnic differences were small and did not reach levels indicating statistical significance
(African American: n=34, 68.0%; White n=92, 67.6%; Latina n=I0l, 71.6%; Asian
American/Pacific Islander n=22, 71.0%; Other n=34, 75.6%).
Mother: Supervised Visits Ordered
Supervised visits were ordered for 152 mothers representing 37.7% of the sample. Racial
and ethnic differences were small and did not reach levels indicating statistical significance
(African American: n=l5, 30.0%; White n=50, 36.8%; Latina n=57, 40.4%%; Asian.
American/Pacific Islander n=, 35.5%; Other n=l9, 42.2%).
Mother: Random Drug/Alcohol Testing Ordered
Racial and ethnic differences in orders for random drug testing were significant (ChiSquare=l 5 .3, p<.01). Random drug/alcohol testing was ordered for 46.1 % of Latinas (65 out of
141) and 44.9% of Whites 61 out of 136), 38% of African Americans (19 out of 50), 37.8% of
mothers classified as belonging to another group (17 out of 45), but 9.7% (3 out of 31) Asian
American/Pacific Islander mothers. Asian American/Pacific Islander mothers, consistent with
lesser indication of substance abuse problems, were less likely to be subjected to testing. Please
see Table 65.

90

Table 65. Mother: Random Drug/Alcohol Testing Ordered

Ordered
Not
Indicated

Total

Black

White

Latino

Asian/PI

Other

165
(40.9%)

19
(38.0%)

61
(44.9%)

65
(46.1%)

3
(9.7%)

17
(37.8%)

238
(59.1%)

31
(62.0%)

75
(55.1%)

76
(53.9%)

28
(90.3%)

28
(62.2%)

Mother: Alcohol Treatment Ordered
Racial and ethnic differences in rates of ordering random drug testing were significant
(Chi-Square=l5.3, p<.01). Alcohol treatment was ordered for 45.4% of Latinas (64 out of 141)
and 39.7% of Whites 54 out of 136), 44% of African Americans (22 out of50), 37.8% of
mothers classified as belonging to another group ( 17 out of 45), but only 12.9% (4 out of 31)
Asian American/Pacific Islander mothers. Asian American/Pacific Islander mothers, consistent
with lesser indication of substance abuse problems, were less likely to be ordered to alcohol
treatment. Please see Table 66.
Table 66. Mother: Alcohol Treatment Ordered

Ordered
Not
Indicated

Total

Black

White

Latino

Asian/PI

Other

161
(40.0%)

22
(44.0%)

54
(39.7%)

64
(45.4%)

4
(12.9%)

17
(37.8%)

242
(60.0%)

28
(56.0%)

82
(60.3%)

77
(54.6%)

27
(87.1%)

28
(62.2%)

Mother: 12-Step Ordered
Participation in a 12-step program was ordered for 108 mothers representing 26.8% of the
sample. Racial and ethnic differences were not statistically significant (African American: n=l2,
24.0%; White n=42, 30.9%; Latina n=39, 27.7%; Asian American/Pacific Islander n=4, 12.9%;
Other n=l 1, 24.4%). As with other substance abuse services Asian American/Pacific Islander
mothers had rates of ordered participation lower than those of other groups. The discrepancy was

91

smaller because 12-step groups were recommended less for all non Asian American/Pacific
Islander mothers than other forms of intervention.
Mother: lndivid ual Counseling Ordered
Individual counseling was ordered for 178 mothers representing 44.2% of the sample.
Racial and ethnic differences did not reach statistically significant levels (African American:
n=l8, 36.0%; White n=65, 47.8. %; Latina n=61, 43.3%; Asian American/Pacific Islander n=13,
41.9%; Other n=21, 46.7%). Mothers in all groups received individual counseling at moderate
levels.
Mother: Other Counseling Ordered
Racial and ethnic differences in whether or not "other" counseling was ordere~ were
statistically significant (Chi-Square=13.6, p>.05). Counseling classified as other- than individual
was recorded as ordered for more than 38.7% (12 out of 31) Asian American/Pacific Islander
mothers, 31.1 % (14 out of 45) "Other" mothers, 29.l % (41 out of141) Latina mothers, 24.3%
(33 out of 136) white mothers but only 8.0% (4 out of50) African American mothers. Other
counseling appears to be a preferred intervention for Asian American/Pacific Islander mothers
who are less likely to be alcohol and drug involved and assigned to substance abuse treatment
than other groups. African American mothers, on the other hand, are rarely assigned to this form
of assistance. Please see Table 67.
Table 67. Mother: Other Counseling Ordered

Ordered
Not
Indicated

Total

Black

White

Latino

Asian/PI

Other

104
(25.8%)

4
(8.0%)

33
(24.3%)

41
(29.1%)

12
(38.7%)

14
(31.1 %)

46
(92.0%)

102
(75.0%)

100
(70.9%)

19
(61.3%)

31
(68.9%)

Mother: Psychological Evaluation Ordered
Psychological evaluation was ordered for 73 mothers representing 18.1 % of the sample.
Racial and ethnic differences did not reach levels indicating statistical significance (African
American: n=7, 14.0%; White n=30, 22.1 %; Latina n=23, 16.3%%; Asian American/Pacific
Islander n=4, 12.9%; Other n=9, 20.0%).
Father: Random Drug Testing Ordered

92

Racial and ethnic differences in rates of ordering random drug testing were significant
(Chi-Square=9.2, p<.05). Random drug/alcohol testing was ordered for 25.01% of Whites (34
out of 136), 24.8% of Hispanics/Latinos (35 out of 141), 17.8% of"Others" (8 out of 45) 12.0%
of African Americans (6 out of 50), and 6.5% of Asian American/Pacific Islander fathers (2 out
of 31 ). Fewer African American fathers than others are married and living with children,
possibly making drug testing seem less important to a resolution of the case. There is less drug
involvement in Asian American/Pacific I.slander families. Please see Table 67.
Table 67. Father: Random Drug Testing Ordered

Ordered
Not
Indicated

Total

Black

White

Latino

Asian/PI

Other

85
(21.1%)

6
(12.0%)

34
(25.0%)

35
(24.8%)

2
(6.5%)

8
(17.8%)

318
(78.9%)

44
(88.0%)

102
(75.0%)

106
(75.2%)

29
(93.5%)

37
(82.2%)

Father: Alcohol Abuse Treatment Ordered
Racial and ethnic differences in rates of ordering random drug testing were significant
(Chi-Square=l0.1, p<.05). Alcohol abuse treatment was ordered for 24.1 % of Hispanics/Latinos
(34 out of 141) and 22.8% of Whites 31 out of 136), 12% of African Americans (6 out of 50),
11.1% of"Others" (5 out of50), and 6.5% of Asian American/Pacific Islander fathers (2 out of
31 ). Please see Table 68.
Table 68. Father: Alcohol Abuse Treatment Ordered

Ordered
Not
Indicated

Total

Black

White

Latino

Asian/PI

Other

78
(19.4%)

6
(12.0%)

31
(22.8%)

34
(24.1%)

2
(6.5%)

5
(11/1%)

325
(80.6%)

44
(88.0%)

105
(77.2%)

107
(75.9%)

29
(93.5%)

40
(88.9%)

Father: Individual Counseling Ordered
Racial and ethnic differences in individual counseling were statistically significant (ChiSquare= 9.8, p<.05). Individual counseling was recorded as ordered for 29% of Asian

93

American/Pacific Islander fathers (9 out of 31 ), 24.3% of Whites (33 out of 136), 20% of
"Others" (9 out of 45), 17. 7% of Hispanics/Latinos (25 out of 141 ), and 6.0% of African
American fathers (3 out of 50). Please see table 69.
Table 69. Father: Individual Counseling Ordered

Ordered
Not
Indicated

Total

Black

White

Latino

Asian/PI

Other

79
(19.6%)

3
(6.0%)

33
(24.3%)

25
(17.7%)

9
(29.0%)

9
(20.0%)

324
(80.4%)

47
(94.0%)

103
(75.7%)

116
(82.3%)

22
(71.0%)

36
(80.0%)

Father: Other Counseling Ordered
Other-than individual counseling was ordered for 57 fathers representing 14.1 % of the
sample. Racial and ethnic differences did not reach levels indicating statistical significance
(African American: n=2, 4.0%; White n=23, 16.9%; Hispanic/Latino n=l9, 13.5%; Asian
American/Pacific Islander n=7, 22.6%; Other n=6, 13.3%).

94

4. Hearings and Case Changes by Ethnic Group
The data in this section are derived from the case record review database.
From the 403 case record reviews, information about the number and type of hearings
was gathered as well as the type of changes that occurred for the child throughout the case.
Number of Court Hearings Over the Course of the Case By Ethnicity

The number ofhearings convened during the course of the case was related to child's
ethnicity. African Americans appeared to have been involved in more hearings on average (9. 34,
sd-= 8.11). Whites were involved in an average of 6.31 hearings (sd = 4.92), Hispanics/Latinos
6.57 hearings (sd = 5.21), Asian American/Pacific Islanders 6.71 hearings (sd = 4.20), and
Others 5.60 hearings (sd = 4.40). Please see Table 70.
'

Table 70: Number of Court Hearings Over the Course of the Case By Ethnicity
Hearings
Avg.
(sd)

Total

Black

White

Latino

Asian/Pl

Other

6.73
(5.48)

9.34
(8.11)

6.31
(4.92)

6.57
(5.21)

6.71
(4.20)

5.60
(4.40)

To help summarize this information and analyze the hearings and case changes regarding
children in the CWS, three descriptive categories were created. The first category was Placement
and Placement Related Changes, which included specific change in status and placement (e.g.,
FM, FR, Bypass, PP), custody, runaways, emancipations, and dismissals. The second was
Service Related Changes, which included visitation rights, orders for counseling, family services,
treatment, other referrals, and compliance with service plan. The third category was Psychosocial
Status Related changes, which included changes in health, mental health, disability, education,
economic, criminal, substance abuse, family violence, cultural, and other adjustment conditions.
Within each of these 3 categories, changes were coded positive, negative, or other. Others are
changes which could not be determined as either positive or negative, for example "mother got
married to man not the father" could possibly be positive if this created a more stable home
environment, or possibly negative if this created more disruption in the home given a tenuous
relationship between the child and the new father. There were no ethnic differences in

Placement/Placement Related, Services, or Psychosocial Status related changes by ethnicity.
Placement Related Changes

There were no differences in placement related changes by ethnicity. The average
number of positive placement related changes occurring over the course of the case were 1.32
changes (sd = 0.99). African Americans were involved in an average of 1.46 positive

95

changes (sd = 1.07), Whites 1.26 (sd = 0.89), Hispanics/Latinos 1.38 (sd = 1.05), Asian
American/Pacific Islanders 1.23 (sd = 0.76), and Others 1.22 positive changes (sd = 1.15). Please
see Table 71.
Table 71: Number of Positive Placement Related Changes
Positive
Placement
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

1.32
(0.99)

1.46
(1.07)

1.26
(0.89)

1.38
(1.05)

1.23
(0.76)

1.22
(1.15)

The average number of negative placement ·related changes occurring over the course of
the case were 0.63 changes (sd = 1.29). African Americans were involved in an average of 0.55
negative changes (sd = 1.27), Whites 0.63 (sd = 1.27), Hispanics/Latinos 0.70 (sd = 1.46), Asian
American/Pacific Islanders 0.42 (sd = 0.76), and Others 0.62 negative changes (sd = 1.13).
Please see Table 73.
Table 73: Number of Negative Placement Related Changes
Negative
Placement
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

0.63
(1.29)

0.54
(1.27)

0.63
(1.27)

0.70
(1.46)

0.42
(0.76)

0.62
(1.13)

The average number of other placement related changes occurring over the course of the
case were 0.63 changes (sd = 0.91). African Americans were involved in an average of0.78
other changes (sd = 0.91), Whites 0.68 (sd = 0.94), Hispanics/Latinos 0.54 (sd = 0.88), Asian
American/Pacific Islanders 0.45 (sd = 0.68), and Others 0.71 other changes (sd = 1.03). Please
see Table 74.

96

Table 74: Number of Other Placement Related Changes
Other
Placement
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

0.63
(0.91)

0.78
(0.91)

0.68
(0.94)

0.55
(0.88)

0.45
(0.68)

0.71
(1.04)

Service Related Changes
There were no differences in service related changes by ethnicity. The average number of
positive service related changes occurring over the course of the case were 1.49 changes (sd =
1.94). African Americans were involved in an average of 1. 72 positive changes (sd = 2.15),
Whites 1.46 (sd = 1.89), Hispanics/Latinos 1.57 (sd = 2.16), Asian American/Pacific Islanders
1.23 (sd = 1.43), and Others 1.29 positive changes (sd = 1.94). Please see Table 75.
Table 75: Number of Positive Service Related Changes
Positive
Service
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

1.49
(1.94)

1.72
(2.15)

1.46
(1.89)

1.57
(2.16)

1.23
(1.43)

1.29
(1.94)

The average number of negative service related changes occurring over the course of the
case were 0.07 changes (sd = 0.32). African Americans were involved in an average of 0.02
negative changes (sd = 0.14), Whites 0.09 (sd = 0.37), Hispanics/Latinos 0.06 (sd = 0.31), Asian
American/Pacific Islanders 0.16 (sd = 0.45), and Others 0.04 negative changes (sd = 0.21 ).
Please see Table 76.

97

Table 76: Number of Negative Service Related Changes
Negative
Service
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/Pl

Other

0.07
(0.32)

0.02
(0.14)

0.09
(0.37)

0.06
(0.31)

0.16
(0.45)

0.04
(0.21)

The average number of other service related changes occurring over the course of the
case were 0.06 changes (sd = 0.25). African Americans were involved in an average of0.12
other service related changes (sd = 0.52). Changes for other ethnic groups were as follows:
Whites 0.04 (sd = 0.19), Hispanics/Latinos 0.08 (sd = 0.32), Asian American/Pacific Islanders
0.03 (sd = 0.18), and Others 0.07 (sd = 0.21). Please see Table 77.
Table 77: Number of Other Service Related Changes
Other
Service
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/Pl

Other

0.06
(0.25)

0.12
(0.52)

0.04
(0.19)

0.08
(0.32)

0.03
(0.18)

0.07
(0.25)

Psychosocial Status Related Changes
There were no ethnic differences in psychosocial status related changes. The average
number of positive psychosocial status related changes occurring over the course of the case
were 2.54 changes (sd = 3.85). African Americans were involved in an average of 3.66 positive
changes (sd = 4.73), Whites 2.39 (sd = 3.36), Hispanics/Latinos 2.40 (sd = 3.64), Asian
American/Pacific Islanders 2.97 (sd = 5.55), and Others 1.93 positive changes (sd = 3.26). Please
see Table 78.

98

Table 78: Number of Positive Psychosocial Status Related Changes
Positive
Psychosocial
Status
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

2.54
(3.85)

3.66
(4.73)

2.39
(3.36)

2.40
(3.64)

2.97
(5.55)

1.93
(3.26)

The average number of negative psychosocial status related changes occurring over the
course of the case were 1.93 changes (sd = 3.31). African Americans were involved in an
average of2.68 negative changes (sd = 4.42}, Whites 2.07 (sd = 3.35), Hispanics/Latinos 1.82
(sd = 3.23}, Asian American/Pacific Islanders 1.48 (sd = 2.61), and Others 1.31 negative changes
(sd = 2.15). Please see Table 79.

Table 79: Number of Negative Psychosocial Status Related Changes
Negative
Psychosocial
Status
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/PI

Other

1.93
(3.31)

2.68
(4.42)

2.07
(3.35)

1.82
(3.23)

1.48
(2.61)

1.31
(2.15)

The average number of other psychosocial status related changes occurring over the
course of the case were 0.46 changes (sd = 1.01). African Americans were involved in an
average of 0.50 other changes (sd = 1.05), Whites 0.41 (sd = 1.02), Hispanics/Latinos 0.47 (sd =
1.10), Asian American/Pacific Islanders 0.52 (sd = 1.00), and Others 0.47 other changes (sd =
0.66). Please see Table 80.

99

Table 80: Number of Other Psychosocial Status Related Changes

Other
Psychosocial
Status
Service
Related
Changes
Avg.
(sd)

Total

Black

White

Latino

Asian/Pl

Other

0.46
(1.01)

0.50
(1.05)

0.41
(1.02)

0.47
(1.10)

0.52
(1.00)

0.47
(0.66)

100

VIII. Key Informant Interviews
The qualitative component ofthis study was designed with input from the project
advisory board. The aim ofthis component of the study was to explore the perceived dynamics,
factors, policies, and possible solutions associated with the over-representation ofchildren of
color in the child welfare system.
Methods
Eight in-depth interviews, of approximately one hour in length, were conducted with key
informants in management positions in DFCS. Interviewees responded to six (6) open ended
questions that explored their overall perspective on the presence of children of color in their
particular unit/region, which groups were under/over represented, the factors related to this
disproportionate representation, and current federal, state or agency policies that they thought
contributed to under/overrepresentation. In addition, interviewees were invited to describe any
suggestions for ways in which they thought the system could respond to over/underrepresentation. Demographic information was collected at the close of the interview.
Sample
Key informants were selected based on their overall experience in child welfare and
capacity to address the representation of children of color at specific choice points in different
parts of the child welfare system. Two interviewees from South County, where a vertical case
management model is employed, provided qualitative data across different facets of the child
welfare system. Six interviewees were specifically associated with each of the units listed below:
•
•
•
•
•
•

Emergency Response (ER)
Dependent Intake (DI)
Family maintenance/Family Reunification (FM/FR)
Adoptions
Concurrent Planning
By-Pass of Family Reunification

Interviewees all had extensive experience working in the Department of Social Services.
Years of experience in DSS/Child Welfare Services ranged from 13 to 31 years with an average
of 23 years. One interviewee had a BA in Social Work and seven had earned an MSW degree.
Four of the interviewees were Latino/a, three were Caucasian/Euro American, and one was
African American. Six of the key informants are fluent in speaking another language, primarily
Spanish, and use these language skills in their work.

101

Analysis
Notes were taken during the interview and a content analysis of this data was-conducted to
identify themes that were common across interviews. Specifically, issues and ideas that emerged
in more than one interview were documented in two broad areas: interviewee perception of
over/under-representation of different populations of children in the system and factors attributed
to this over/under-representation. A separate analysis was conducted to summarize and integrate
interviewee suggestions for potential systemic responses to address over-representation of
specific populations in CWS.

Key Informant Interview Findings
Interviewees noted that Santa Clara is a diverse county with a larger percentage of
families of color than are reflected in many other California Counties and that a variety of efforts
to meet the needs of specific populations have evolved over recent years. However, all
interviewees noted that that African American and Hispanic children were dispr9portionately
represented in the Child Welfare System. Several interviewees commented that Asian
American/Pacific Islander and East Indian children are under-represented in the system. Gay,
Lesbian, Bisexual and Transgender families (parents and children) were identified as being
"invisible" in the system. Although interviewees were asked to discuss the representation of
children of color in their specific unit or area, all had extensive experience in other facets of child
welfare that were included in their observations. Several themes related to factors that may

contribute to the over-representation ofchildren ofcolor in the Child Welfare System in Santa
Clara County (and other counties) emerged.from the interview data.
Social and Economic Factors

Interviewees cited poverty, lack of education, unemployment, insufficient job skills, and
lack ofsophistication in navigating through formal systems as contributing factors to the overrepresentation ofAfrican American and Hispanic children in the Child Welfare System (CWS).
Other environmental factors that were deemed to disproportionately impact families ofcolor
included living in communities that are "plagued with drugs and violence. " Poverty was named

as a risk factor for entering CWS. "Families who can afford an attorney and therapist can stay
out of system," one interviewee observed. Nearly all interviewees also identified poverty as a key
ba"ier to exiting the system. Poor families were perceived as less likely to be able to mobilize
resources required to reunify. One interviewee with over twenty years of experience in child.
welfare offered a case in point, "children are not removed solely because of homelessness, but I
have never seen a child returned to a homeless family." The economic downturn and continued
high price of housing in Santa Clara County was identified "making it harder for the have-nots."

Factors related to immigration status were also considered to influence the progress and
outcome for Hispanic/Latino children in CWS. In terms ofentree into the system, cultural norms
for immigrant families, particularly from Mexico, may conflict with dominant social norms
related to raising and disciplining children. "It goes back to how things are done in Mexico,"

102

observed on interviewee, "Some families think nothing of leaving a child alone with an older
child for a few hours." "There are a number of checkpoints in the system," she added, "and we
rely on good diversion programs to keep these families out of the court system." Once in the

system, the barriers for immigrant families are myriad. For example, one interviewee found that
in many instances, "children were adopted by non-relatives because birth mothers were not
married to the father ofthe child and would not provide the fathers name for fear ofgetting the
family n trouble with the INS; as a result, the child was deprived access to relatives on one
side. " The message that the Child Welfare System is not connected to the INS is "not out there
in the community." Another interviewee commented that "If a family is undocumented the
chances of getting a kid back are almost nil because they are in a position where they can't get
section 8 housing, are not eligible for welfare, and have to fake a Social Security Number to get
job."
Individual Bias Based on Race, Class, and Immigration Status

Interviewees consistently identified bias on the part ofsocial workers and other
professionals as an important contributor to the over-representation ofAfrican American and
Hispanic/Latino families in the system. Several interviewees noted that many social workers,
district attorneys, and other professionals apply "an ideal rather than a minimum standard" for
parenting based on their own experience. Moreover, the class, race and country of origin for
professionals in positions of power are, in general, substantially different than that of clients in
the system. Consequently ''there is often bias in the levels of trust accorded to people we see as
radically different."

Substantial concern was expressed over bias among social workers and, specifically that
initial and ongoing assessment "depends on who investigates." One interviewee captured this

theme, "If you were to send three workers out to the same family, particularly if they are from
different cultures, you will get back three different assessments." There was a high value placed
on "staff familiar with different cultural groups who are able to make an appropriate
assessment." Interviewees affirmed the tenet that clear abuse and risk of child safety warranted
involvement in CWS. At the same time, many were concerned that many children, particularly

Black and Hispanic/Latino children were often brought into the system with less compelling
reason than other children. One interviewee questioned substantive consistent differences in the

decisions of specific workers and units: "If you look at data from a worker level, you can see that
some workers file 90 percent of the time and some file infrequently." Another noted that "in
working with an African American family that may have a negative history with the police and
schools, one worker will see the parent as hostile and presume a negative impact on child
development while another might assume the parent had reason to be hostile and would dissipate
the anger." Even in the context of substantiated abuse, "worker perception can impact outcome
and consider family circumstances ...just because more kids of color enter doesn't mean they
have to be freed for adoption."

Historical discrimination and racism was considered by many interviewees to be linked
to poverty and disenfranchisement in families which, in turn, was perceived to increase the

103

likelihood that families would come to the attention of child welfare. Many interviewees noted
that many "children of color are reported on the front end for too many things" largely through
other systems. "People of color are given disproportionate attention from law enforcement, " a
primary source for referrals into the system. "CPS returns a lot of kids in five days," commented
one interviewee. One interviewee reflected a viewpoint common among interviewees, "Police
are more likely to refer families of color, doctors are less likely to report middle class families,
and teachers are more likely to overlook parents who are visible at school."
Systemic Inadequacies and Structural Bias

Remarks such as, "We often demand too much offamilies" and "We don't want to set
families up to fail but we often do" were echoed in a number of interviews. Mandates in multiple
systems that fail to coordinate may particularly impact families of color. "Family members may
have a probation officer, a social worker, and the court all making demands of the family, some
of which may be contradictory," noted one key informant. Another suggested, "Loading up a
service plan looks good in court but it's too much for a client, particularly when the effort to
survive takes 90 percent of day."
Recognition of the extended family system is critical, according to many interviewees,
particularly in relation to working with many communities of color. "Thinking offamily as a

mother-father-child configuration is common but problematic ... it is just not how children are
really raised " CWS often relies on family case conferencing and family decision-making in
addressing the needs of children, particularly in diversion programs. "Resource development
and removal depend on the issues central to the case and how well the extended family can care
for a child," stressed one expert.

Substantial advances have been made in the CWS to me_et the needs ofdiverse families
through culturally specific family centers, efforts to "match" workers and clients in relation to
language and culture, and the development ofhigh quality diversion programs. However, the

CWS is not the sole or even the primary locus for some of the dynamics that impact families of
color, according to some interviewees. For example, DFCS can make a recommendation to not
use bypass, but a District Attorney (DA) representing a child might successfully argue in court
for bypass. "The entire system needs to be reviewed," posited one interviewee, "This is a DA,
community, juvenile justice, and court issue."

Several interviewees suggested that the CWS system in general and the court process in
particular is a "mismatch" for families ofcolor. "Minority groups are intimidated by the system,

don't understand how the system works, and are less likely to speak up in court," summarized
one informant. For example, Hispanic/Latino families, particularly immigrants may have been
socialized to "have respect for education so if their attorney doesn't speak up, they think it's for a
reason." Courts were often described as "rigid," "lacking creativity," or failing to "understand
what is cultural." The form of mandated services is also an important issue. "If a Latino parent is
court ordered to therapy, this may be seen as 'being crazy, while attending a psychoeducational

104

group is less of a problem," observed one respondent. Social workers are often placed in the
position ofeducating both clients and the court ifthey are capable and willing to do so.
Factors Related to Family Characteristics

Substance abuse and domestic violence were identified as factors that frequently bring
families into child welfare. "Drugs fuel a lot ofcases; it's one ofthe primary reasons kids end up
in the system and we need culturally competent substance abuse services and parent education, "
noted one respondent. Another speculated, "If we had enough prevention and treatment, CWS
demand would go down.

A paucity ofsocial services, particularly multi-lingual services, was cited as significant
barriers for many families ofcolor. Interviewees discus_sed the shortage ofsubstance abuse
treatment programs, particularly those geared for women with children and people whose
primary language is not English. Multi-lingual and culturally appropriate domestic violence
services, parenting classes, and other social services were considered in need of development.
The scarcity of accessible services was perceived as "discrimination against poor who have to
rely on free treatment" by more than one informant.
Factors Related to Laws and Policies
The Adoption and Safe Families Act (ASFA) was described as a significant factor in the
overrepresentation of Black and Hispanic/Latino children in CWS. One interviewee

summarized, "AFSA is great when permanent placement outside the home is hands down best
thing for child, the downside is that it is rigid when greater flexibility would be useful." For
example, several informants noted, "adoption by a relative is not a cultural norm." African
American and Hispanic/Latino families may prefer guardianship or other less formal .
arrangements, while "the system pressures for a stronger plan. " One respondent who explained,
"Laws that require permanency against that cultural backdrop mean that children may stay in
the system longer, " summarized the consequences ofthis conflict.
Many interviewees discussed the conflicting timelinesfor reunification in relation to
substance abuse treatment. "Because we have these timelines, they are losing these kids left and

right," observed one respondent who went on to explain, "I push for understating that the
recovery timeline is longer, look at mothers potential for recovery, and try to leave door open for
the parent in permanency planning as it is appropriate." Some respondents suggested that a

substantial number of referrals into the system and concurrent planning cases involved infants
who screened positive for drug exposure at birth, and that Black and Hispanic mothers were
more likely to be impacted by toxicology testing and referral. Several interviewees stressed the

importance of adopting consistent, well-considered practice in relation to implementing
provisions of ASFA in relation to bypass options and concurrent planning. One interviewee
stated, "many bypass decisions are legitimate but other families that are bypassed may deserve a
chance." Another noted that bypass was used more extensively in the first years of
implementation, but that greater individual review had become normative.

105

The Multi-Ethnic Placement Act was identified as both and asset and an impediment in
service children ofcolor. As intended, this measure serves to expedite placement ofchildren into
adoptive homes. As a result, many children are placed cross-racially. According to many

interviewees, some .provisions including the short time period allowed to identify a potential
foster/adoptive home, the option for emergency foster homes to keep a child after 120 days, and
requirements to include county and foster family homes in the pool of prospective placements
results in "random placement with not matching on culture." Even in the context of the law,
many interviewees described making efforts to find a cultural match for children and
encouraging adopting families from different cultures to consider how they would respect their
adoptive child's heritage.

Interviewees also mentioned the California Safety Assessment Tool and Risk Assessment.
Several noted that these instruments are useful in "getting workers on the same page. "At the
same time, they noted that different workers frequently have substantially different results using
the same tool. One interviewee noted that many assessment decisions are made on "spur of the

moment" and do not adequately examine the "whole picture." As a case in point, a worker may
decide a "dirty" drug test is adequate reason to remove a child without assessing whether the
parent attended to the safety needs of the child ("What if when they used, they dropped the child
off at the grandparents?") or even has a serious problem ("What if they are a recreational user?")
Policies related to eligibility for benefits may conflict with the best interests of families.
If a family is eligible for Temporary Assistance for Needy Families (TANF), they loose their
benefits when they lose their child/ren. Section 8 housing eligibility is also jeopardized by
removal of children. Loss of income and housing poses a substantial barrier to reunification for
low-income families.

Interviewee Recommendations
Interviewees acknowledged systemic changes that have improved services to families of
color. "We've come a long way since 73 in having staff representative of clients," stated one
informant. Family resource centers, inclusion ofextended family in decision-making, emerging

models for sharing resources and decision making with communities were all lauded as
important improvements. At the same time, several interviewees noted that "it's still a problem

that needs to be addressed" and that issues and data related to overrepresentation of children of
color in child welfare continue to be championed by concerned staff. Interviewees suggested that
efforts should be continued and evaluation. Key informants outlined potential responses,
building on current programs and procedures.
Approaches to Working with Families
•
•

Continue approaches that "work" including family case conferencing, intensive family
maintenance, and use of psychoeducational groups.
Expand outreach for the Family Resource Centers.

106

•
•
•

Explore the adoption or adaptation of vertical case management in other areas of the county
(which appears to be associated with fewer bypass decisions and higher rates of family
reunification in shorter time periods).
Support development of peer counseling for birth parents (e.g., pay mentors who have
successfully reunified to serve as counselors).
Continue development of"family to family," a new program that includes community
members in decision making.

Training for Social Workers
•
•

Train social workers that make first line decisions to be able to reflect on their own practice.
Educate social workers about different cultural groups and mandate that workers continue to
get training.
Provide training that operationalizes "minimal standards" and minimizes staff imposition of
their own personal experience, frameworks, and values.

Training Outside of the Child Welfare System
•
•

Reach Latino families through Latino media; disseminate information about services and the
fact that Social Services does not report to immigration.
Educate adoptive families about importance of supporting their child's sense of heritage.

Investment in Prevention and Early Intervention
•
•
•
•
•

Seek allocation of federal and state funds for diversion and prevention services (that are
currently not covered and are at risk because of current economic conditions). "Prevention
should be funded equally with out of home funding," specified one interviewee.
Explore avenues to claim funds for social workers in school for prevention.
Expand use ofmentorship (e.g., holding groups with parents newly involved in CPS to tutor
them about the about system and the court process.
Explore ways to provide education for recent immigrants on legal issues related to families
and child welfare.
Expand voluntary services (so that families at risk receive needed services and are not "raked
over the system.")

Agency Policy
•
•

Provide mechanisms for supervisors to track data that may reflect bias on the individual
worker level and increase worker accountability for meeting definitions in structured
decision-making.
Develop and implement guidelines that provide debt relief. "Once families are in the system,
it is costly to get out, for example, they are charged for foster care. Legislation was passed to
address this, but guidelines. are not yet developed.

107

•

Structure worker time to account for the fact that Spanish-speaking cases (and other
immigrant cases) require more time.

Facilitation of Collaboration Between Systems
•
•
•
•

Enhance collaboration between child welfare, substance abuse, and other fields impacted by
conflicting "timelines."
Facilitate greater collaboration between family court and child welfare; allocate resources for
services through family court that may reduce "spiraling" into the dependency system.
Facilitate an increase in multilingual services in substance abuse, DV, family service
agencies, and youth services (such as big brothers/big sisters).
Work to institute a 6-month moratorium on loss of section 8 housing for families that appear
to be on a reunification track.

108

IX. Summary and Conclusions
Pathways Through the Child Welfare System: Summary Descriptions by
Ethnic Group and Implications for the Child Welfare System
The significant case differences related to ethnicity described in parts VII of this report,
as well as the analysis of the closed case sample enabled us to identify key variables related to
each ethnic group involved with the CWS. The following presents summary descriptions of
pathways through the CWS for each of the four racial/ethnic groups that we focused on in Phase
2. Individual and family-related characteristics, as well as system-related factors and services
are also included in the descriptions. In a few instances variables are included that failed to
reach significance in our analyses of the case record review sample, but were significant in the
larger analysis of the overall closed case sample.
Figure 2 provides a summary chart of differences in pathways across the four
racial/ethnic groups and related individual, family and system characteristics.
African American
Child and Family Characteristics
At 14.7%, African American children represent the second smallest group in the county's
CWS, but are overrepresented given that they represent only 4% of the general child population
on Santa Clara county. According to our case review sample, at entry into the CWS African
American children appear to be younger than children in other ethnic groups. African American
children are less likely than either White or Latino children to have mental health or behavioral
problems and almost all do not require a translator in court proceedings.
African American mothers tend to be younger than mothers in any of the other ethnic
groups, they also tend to be single or never married, and almost none are living with a spouse.
African American mothers' relatively young age, and their likelihood of being a single parent
may increase their risk of experiencing parenting problems--possibly due to less maturity and
experience in everyday life matters. This theory is corroborated by the finding that mothers alone
represent the largest proportion ofperpetrators within this ethnic group. More African
American mothers are welfare eligible, and have more children in the household, as well as
more children involved in the CWS than average-all of these factors have been shown in the
research literature to increase the likelihood of CWS involvement. Despite these stressors,
African American mothers tend to have a lower than average rate ofreported mental health
problems. Yet, African American mothers do tend to have a higher than average rate of
substance abuse, and higher than average rates ofcriminal justice system involvement at the
time their child(ren) enter the CWS. African American mothers also are likely to have a high
rate ofpast criminal history, and previous incarceration, as well as a relatively high rate of
criminal episodes. Domestic violence involvement is relatively high among this group, as well.

109

System-Related Factors
CWS cases from this group also appear likely to be initially reported by law enforcement
and legal professionals, as well as personnel at medical sites. For African American children in
this sample, the most frequently occurring initial reason for removal from the home was "other
maltreatment," although general or severe neglect and physical abuse also appear to be
somewhat common.
Once in the CWS, African American children have a higher than average rate ofbeing
assigned to family reunification services after the jurisdictional/dispositional hearing, and they
most often experience their initial out-of-home placement with a relative. Regarding services,
African American children appear less likely than other ethnic groups to be ordered to attend
counseling services (both individual and other), and they also appear less likely to have a
psychological evaluation ordered as part of their initial service plan-this result appears
consistent with the finding that African American children tend to have lower rates of mental
health and behavioral problems.
Like other ethnic groups, African American mothers are often ordered to attend parent
education services. It is notable that African American mothers appear to have a relatively low
rate of being ordered to complete random drug testing, and 12-Step services-although they do
tend to be frequently ordered alcohol treatment. African American mothers also seem to not be
ordered individual or other counseling services. African American fathers in the CWS tend to
receive fewer ordered services than other ethnic groups-possibly because fewer African
American fathers are involved in their children's life at the time of entrance into the CWS, thus
making services to the father possibly less important to the resolution of the case.
In general, African American children have more court hearings over the course of their
case than other ethnic groups-perhaps due to the fact that their cases tend to be lengthier than
other ethnic groups. African American children also appear to have a higher than average rate of
being removed from their family within their current CWS episode-possibly indicating that
family instability and risk factors associated with their entrance into the CWS persist throughout
their case history. African American children appear to have longer than average stays in each
out-of-home placement than other ethnic groups-a result that is consistent with the finding that
African American children have a longer than average total amount of time in out-of-home
placement, a longer than average total case duration, and the finding that they have more court
hearing than any other ethnic group. The mostfrequently occurring last out-of-home placement
for African American children tends to be a relative home or a foster family home. At case
closure, African American children have a higher than average rate of being in permanent
placement services. It is notable that African American children tend to have a very low rate of
family reunification services at case closure, perhaps indicating that once removed from their
family, African American children tend not to return to their family- a finding corroborated by
comments of focus group participants during Phase I of this study.

110

Implications for Intervention with African American Children and Families in the Child
Welfare System
Programs targeting this group should consider these factors, and perhaps, especially the
psychosocial background ofthe mother. There are several factors related to being a young, single
parent in a household with many children that can make effective child rearing especially
difficult. In addition, African American mothers appear to be especially isolated The lack of
psychosocial support coupled with financial difficulties can jeopardize healthy parenting
behavior. Additionally, the comparatively high rates ofsubstance abuse and criminal justice
system involvement among African American mothers may also interfere with their ability to
parent effectively.
Given the multitude ofstressors facing African American mothers, the finding that they
have a lower than average rate ofmental health problems appears somewhat contradictory. This
finding may be interpreted in many ways. Perhaps the relatively low rate of mental health
problems among African American mothers suggests that criminal justice system involvement
and/or substance abuse may mask mental health problems in this population. Alternatively,
African American mothers in the CWS may be viewed as closer to a stereotypical view of
African American families held by officials and professionals-and thus they may be viewed as
less deviant and dysfunctional. Still another view is that African American mothers actually do
have fewer mental health problems, and if this is the case, it represents an important protective
factor that may increase the effectiveness of services.
Of concern are the relatively low rates ofrandom drug testing and 12-step services
orderedfor African American mothers--especially in light of their increased likelihood of having
a substance abuse problem. Of even greater concern is the relative lack ofservices provided to
African American families and their children. Given that African American children have the
lengthiest stays in out-of-home placement and in the CWS in general, more services that are
targeted toward substance abuse, as well as the social isolation and financial hardship of
African American mothers are clearly warranted. Parenting education, alone may not be
sufficient to meet the needs of these highly vulnerable and isolated families. Indeed,parenting
education that fails to address the cultural and psychosocial needs of this population may
actually be counterproductive (Repetti et al., 2002).
Research has indicated that programs ofhome visitation that promote positive healthrelated behaviors in mothers ofyoung children, competent care oftheir children and linkage
with needed health care and human services, reduce rates of criminality, problems related to
substance abuse and child abuse and neglect among young, unmarried, isolated poor mothers
(Olds et al., 1997, 1998). Studies that follow children of mothers involved in such home
visitation programs, into adolescence have found these youth have fewer serious behavioral and
adjustment problems than youth whose mothers did not participate in home visitation programs.
Such programs may be a more effective way than parenting education to address the needs of
these vulnerable African American families.

111

White
Child and Family Characteristics
While White children comprise the largest segment of Santa Clara county's child
population (45%), at 25.8%, White children constitute the second largest group in the county's
child welfare population. According to our case review sample, they appear to be older than
average. They also seem more likely than children in other ethnic groups to have a mental
illness or behavioral problem and they have little need for a translator in court proceedings.
White mothers tend to be older than average; they are often single due to divorce,
separation, or death ofa spouse. High proportions of White mothers are suffering from mental
health problems, corroborated by a relatively high proportion of them being involved in the
mental health system. White mothers appear likely to have a high school level education or
above, and they generally have a lower than average number of children in their household. They
also appear more financially secure than other ethnic groups and have a comparatively low
frequency of welfare eligibility. White mothers tend to have a lower than average rate of
involvement in the criminal justice system, as well as fewer criminal episodes, and a lower than
average rate of incarceration. However, this group had the highest rate of involvement in
domestic violence.
System-Related Factors
Law enforcement and legal professionals commonly report the maltreatment of White
children. White children tend to be initially removed from their home for "other maltreatment, "
although physical abuse is also somewhat common.
Once in the CWS, White children are often assigned to family reunification services as a
result ofthe jurisdictional/dispositional hearing, and their initial out-of-home placement is most
likely to be with a relative. White children are more likely than any other ethnic group to be
ordered to attend individual counseling, and they also tend to be ordered to participate in other
counseling services. Additionally, White children are more likely than any other ethnic group to
be ordered to complete a psychological evaluation. These findings, in conjunction with the
relatively high rates of maternal mental illness, suggest that White families in the CWS may
experience an increased likelihood ofmental health problems.
Like other ethnic groups, White mothers are often ordered to attend parent education
services. They appear to be ordered substance abuse services at a rate consistent with the
average. White mothers seem to have a higher than average rate of being ordered to attend
individual counseling-a finding that is consistent with their higher than average mental health
problems. White fathers tend to have a slightly higher than average rate of being ordered both
substance abuse and counseling services.

112

In general, White children have a slightly lower than average number oftimes they are
removed from their family within their current CWS episode. They tend to have a shorter than
average stay in each out-of-home placement-a result that is consistent with the finding that
White children have a lower than average total length of time in out-of-home placement, and a
lower than average total case length. The mostfrequently occurring last out-of-home placement
for White children tends to be a relative home. At case closure, White children are most likely to
be in family maintenance services; they have a lower than average rate of being in permanent
placement services.

Implications for Intervention with White Children and Families in the Child Welfare System

Programs targeting this group should consider these factors, especially the vulnerability
of the child and mother to mental or behavior problems. Mental illness and single parenting may

make effective child rearing difficult for White families and children with mental health and
behavioral problems may be difficult to parent. It is especially important to note that this group
had the highest rate of domestic violence involvement. Parent education programs might
consider these factors in their service approach.

Additionally, a comparatively low proportion of White mothers are eligible for welfareand their relative financial stability may be viewed as an important protective factor in their short
duration of time in the CWS. Also, a relatively high proportion of White mothers appear to have
at least a high school education, indicating their ability to benefit from intervention approaches
designed for the mainstream population. These findings suggest that White mothers may be in a

better position than other ethnic groups to benefit from available child welfare services, that the
available services are targeted towards a more mainstream population, or that White families
are simply offered more services-again corroborating the finding that White children have
comparatively short stays in the CWS.
Hispanic/Latino
Child and Family Characteristics
Latino children constitute 30% of the general child population in Santa Clara county and
at 53.5% are vastly overrepresented in the county's CWS. Latino children tend to have a higher

than average rate of being non-English speaking, and to need a translator in court proceedings.

Latino children in the CWS are more likely to be female and have a low occurrence of
behavioral problems.

Approximately 18% ofLatina mothers need translation services in court proceedings,
25.5% were born outside of the United State and this group also appears to be relatively
undereducated Many Latina mothers also have more children than average, and they often
appear to be in a living situation without a spouse. Additionally,financia/ difficulties are
reflected in Latina mothers' tendency to be welfare eligible. For Latina mothers, there is also
indication of substance abuse problems, as well as criminality, as indicated by the high

113

percentage that are involved with the criminal justice system. Similar to other ethnic minority
mothers, there are factors related to single parenting in a household with many children, making
effective child rearing especially difficult. Compounding this situation is their status as
immigrants and their need for translators in court proceedings. Like African Americans and
Whiles, Latinos have high rates ofdomestic violence involvement.
System-Related Factors
CWS cases from this group appear likely to be reported by school personnel. For Latino
children, the most frequently occurring reason for removal from the home was "other
maltreatment," although they also seem to have higher than average rates ofgeneral neglect and
physical abuse.
·
Once in the CWS, Latino families tend to be assigned to family reunification services,
and their initial-out-ofhome placement is most often with a relative. Latino children appear to
be ordered to c,l!tend individual and other counseling at a rate consistent with the average.
Like other ethnic groups, Latina mothers are frequently ordered to attend parent
education services. They are also ordered to attend substance abuse services, including random
testing, alcohol treatment and 12-Step at slightly higher than average rates-a result that is
consistent with their relatively high rate ofsubstance abuse. Latino fathers also seem to be
ordered random testing and alcohol treatment at slightly higher than average rates-perhaps
indicating Latino families' stressors may be compounded by parental substance abuse by both
mothers and fathers.
Latino children appear to have shorter average stays in each out-ofhome placement than
African Americans or Asian American/Pacific Islanders-this is consistent with their shorter
total length oftime in out-ofhome placement and total length of their case. Datafrom our
overall sample analysis indicated that Latino children also experienced a relatively high number
of unique placement homes-although they had lower than average stays per placement,
suggesting that these children have multiple placements and relatively short stays in each
placement. The most frequently occurring last out-ofhome placement for Latino children tends
to be a relative home or the Children 's Shelter. They also seem to have a higher than average
rate ofhaving their case closed with permanent placement services, and a lower than average
rate of ending their case with family maintenance services, perhaps suggesting that like African
American children, once they are removed from their home, Latino children are not likely to
return.

Implications for Intervention with Latino Children and Families in the Child WeIfare System

114

Programs targeted to Latino families in the CWS should consider these factors-especially acculturation issues, differences in cultural values and the language barriers often
experienced by Latino families. Effective services for Latino families should be culturally
sensitive and conducted in the participants' primary language. Yet, the problem of offering

effective services may be compounded by the fact that a relatively high proportion of Latina
mothers have less than a high school education, and they also have a slightly higher than average
rate of welfare eligibility. Indeed, low educational attainment and financial difficulties may make

effective access and utilization ofservices far less likely.

Substance abuse and criminal justice system involvement also impact Latina mothers,
and Latino fathers have a comparatively high rate of being ordered to attend substance abuse
services. These trends suggest that Latino families not only have cultural, financial and

educational barriers to services, but that they also have additional psychosocial stressors that
should be considered. Effective services with Latino families should consider all of the myriad

stressors affecting this population.

Additionally, like African American families, the relatively low rate of mental health
problems and involvement in the mental health system appear contradictory to the multitude of
stressorsfacing Latina mothers. In addition to the possible reasons listed above, Latina mothers
may be under-diagnosed by mental health professionals due to cultural and language barriers.
CWS services might consider the possibility of undiagnosed mental illness in this population.
Asian American/Pacific Islander
Child and Family Characteristics
At 5 .1 %, Asian American /Pacific Islander children represent the smallest group in the
county's CWS (not including children of"Other" ethnicities), and as they constitute 21 % of the
general child population in Santa Clara county are underrepresented in the CWS. Our analysis of
the overall sample indicated that Asian American/Pacific Islander children are more likely than
the other three groups to be assigned to voluntary family maintenance services (VFM). The

description that follows, focuses on Asian American /Pacific Islander families that receive
involuntary services from the CWS.

Similar to Latinos, Asian American/Pacific Islander children are often non-English
speakers who require translation services in court proceedings. However, compared to Latinos,
Asian American/Pacific Islander children have a much higher rate of being non-English
speakers. Asian American/Pacific Islander children also appear to be slightly younger than the
average and they have a relatively low rate of mental health problems.
Like their children, Asian American/Pacific Islander mothers tend to be non-English

speaking and also have a very high rate of immigrant status. A relatively high proportion of
Asian American/Pacific Islander mothers appear to be married and/or in a living situation with
a spouse. These mothers also tend to be older than the average; they have/ewer children in

115

their household and tend to have a low education level. The mothers are also less likely than
African American and Latino mothers, but more likely than White to be welfare eligible. Asian
American/Pacific Islander mothers seem to have relatively few substance abuse and criminal
problems, although they do tend to have higher than average rates of mental health problems,
and involvement in the mental health system. Issues surrounding mental health may be related to
acculturation and the challenges of adjusting to mainstream American society. Asian
American/Pacific Islander fathers are comparatively more often reported as perpetrators,
possibly indicating that although Asian/PI families are relatively intact, fathers, as well as
mothers, have difficulty with parenting behaviors that are acceptable in mainstream American
society. Domestic violence involvement is high, but lower than the three other groups.
System-Related Factors
CWS cases from this group are reported by personnel at a school site at a higher than
average rate. Asian American/Pacific Islander children appear to be most frequently removed
from their home due to physical abuse, and comparatively, Asian American/Pacific Islander
children have the highest rate ofphysical abuse than any other ethnic group.
Once in the CWS, Asian American/Pacific Islander children tend to receive family
reunification services as a result ofthe jurisdictional/dispositional hearing-although they do
receive family maintenance services at a higher rate than any other ethnic group-possibly
suggesting that family maintenance services may be viewed by the CWS as more effective in
two-parent households, or more effective with this ethnic group. Asian American/Pacific
Islander children often experience their initial out-of-home placement at a foster family home.
Asian American/Pacific Islander children tend to be ordered other counseling at a higher rate
than other ethnic groups.
Like other ethnic groups, Asian American/Pacific Islander mothers are often ordered to
attend parent education services. Asian American/Pacific Islander mothers appear to have low
rates of being ordered to attend substance abuse services, such as random testing, alcohol
treatment and 12-Step-a result that is consistent with the finding that these mothers appear to
have a relatively low rate of substance abuse problems. Asian American/Pacific .Islander mothers
are ordered to receive individual counseling at a rate higher than that of other groups. Asian
American/Pacific Islander fathers appear to have comparatively high rates of being ordered to
attend both individual and other counseling These result are consistent with the finding that
Asian American/Pacific Islander mothers have higher rates of mental health problems and Asian
American/Pacific Islander fathers are often the perpetrators of the initial maltreatment, thus
possibility suggesting a higher incidence of actual or perceived behavioral dysfunction.
In general, Asian American/Pacific Islander children have a lower than average number
of times they are removedfrom their families in their current episode. Asian American/Pacific
Islander children in the case record review sample of non-VFM cases, appear to have lengthier
than average stays in each out-of-home placement-a result that is consistent with the finding
that these children have longer than average total time in out-of-home placement, as well as a

116

longer than average total case. The most frequently occurring.final out-of-home placement/or
Asian American/Pacific Islander children tends to be a foster family home, and they tend to have
their case closed with permanent placement services.
Implications for Intervention with Asian American/Pacific Islander Children and Families in
the Child Welfare System

Prevention and intervention services for Asian Americarz/Pacific Islander families should
not only be culturally appropriate, taking into account acculturation and differences in cultural
values, but they also need to be conductedin the family's native language. Additionally, given
the comparatively high rate ofmarried parents in this group, and the finding that Asian
Americarz/Pacific Islander fathers tend to be the perpetrators ofphysical abuse, programs for
this group also need to target the whole family. However, the problem of offering effective
services may be compounded by the fact that a relatively high proportion ofAsian
Americarz/Pacific Islander mothers have less than a high school education.
Although Asian Americarz/Pacific Islander mothers are less likely to have a substance
abuse problem, their comparatively high rate ofmental health problems and involvement in the
mental health system should be taken into account when targeting services to this group.

Stressors related to acculturation may increase the risk of mental health problems in this
population-again underscoring the need to culturally appropriate and specific services for Asian
American/Pacific Islander families.

Asian Americarz/Pacific Islander families also appear to have certain protective factors
that may make culturally appropriate services with this group more effective. For instance,
Asian Americarz/Pacific Islander mothers tend to have low rates ofsubstance abuse, criminal
justice system involvement, incarceration and welfare eligibility, and these families in general
have fewer children and fewer siblings involved in the CWS. These protective factors may be
utilized to increase the effectiveness of services with this population. At the same time the usual
array ofchild welfare services may not meet the needs of this population.

117

Figure 2. Chart of Characteristics by Ethnic Group

White

Latino

Black

Asian/Pacific Islander

Child

Child

Child

Child

•

•

•

•
•
•

•
•

53.5% of SC CWS (April
2002)
46.4% closed case sample
35.0% case record review
sample

t Females
t Need for translator in court
t Immigrant status

•
•

t

25.8% of SC CWS (April
2002)
29 .5% closed case sample
33.7% case record review
sample
Age

-l- Need for translator in court

14.7% SC CWS (April
2002)
11.1 % closed case sample
12.4% case record review
sample

•
•

J.. Age

Mental health problems

-l- Need for translator in court
-l- Mental health problems

Behavioral problems

-l- Behavioral problems

J.. Behavioral problems

t
t

Mother

Mother

Mother

Age

J.. Age

J..Age

t Not living with spouse
t Need for translator in court
t Number of children in the
household

J.. Education level

t
t
t
t

Welfare eligibility
Domestic Violence
Substance abuse problems

Criminal justice system
involvement

J.. Mental health problems

t
t

Single, divorced

-l- Need for translator in court
-l- Number of children in the

household

t

t

Single, never married

-l- Need for translator in court

t

Number of children involved
in the CWS

-l- Criminal justice system

t
t
t
t
t

t

-l- Mental health problems

Education level

-l- Welfare eligibility

t
t

Domestic Violence
Substance abuse problems

involvement

Mental health problems

Education level
Welfare eligibility
Domestic Violence
Substance abuse problems

Criminal justice system
involvement

5.1% SC CWS (April 2002)
9.3% closed case sample
7.7% case record review
sample

J.. Age

t
t

Need for translator in court
immigrant status

-l- Mental health problems
Mother

t Age
t Married, living with spouse
t Need for translator in court
-l- Number of children in the
household

-l- Education level

t Welfare eligibility
t Domestic Violence
-l- Substance abuse
-l- Criminal justice system
involvement

t

Mental health problems

118

Figure 2. Chart of Characteristics by Ethnic Group (Continued)
Latino

White

Black

Asian/Pacific Islander

System-Related Factors

System-Related Factors

System-Related Factors

System-Related Factors

tGeneral ne~lect by mother
and others, a so physical abuse,
and other

t "Other" maltreatment or
thhsical abuse by mother,
at er and/or others

t General or severe neglect by
mother only, and physical
abuse, and other

t Physical abuse by fathers

t Initial reporter law
enforcement or legal
professional, also school site
personnel

t Initial reporter law
enforcement or legal
professional

t Initial reporter law
enforcement or legal
professional, also medical site
personnel

t Ass~nment to family
reuni 1cation services

t Ass~nment to family
reuni 1cation services

t Ass~nment to family
reuni 1cation services

t Initial out-of-home
placement with relative, or at
Children's Shelter

tinitial out-of-home
placement with relative, or
Children's Shelter

tinitial out-of-home placement
Children's Shelter or with
relative

t Mothers ordered to complete
parent education services

t Mothers ordered to complete
parent education services

t Mothers ordered to complete
parent education services

t Mothers ordered to complete
substance abuse services

t Mothers ordered to complete
counseling services

,J, Mothers ordered to complete

,J, Mothers ordered to complete
counseling services

,J, Child ordered to attend

counseling

t Child ordered to attend
counseling

,J, Child ordered to attend

counseling

t Child ordered to attend
counseling

t Last out-of-home placement
relative home, or Children's
Shelter

t Last out-of-home placement
relative home

t Last out-of-home placement
relative home-or foster family
home

t Last out-of-home placement
a foster family home

,J, Average stay in each out-of-

home placement

home placement

,J, Average stay in each out-of-

counseling services ·

t Average stay in each out-ofhome placement (longest)

,J, Length of time in out-ofhome placement in general

,J, Length of time in out-ofhome placement in general
(shortest)

t Case bein~ closed with
permanent p acement services

t Case being closed with
family maintenance services

t Case bein~ closed with
permanent p acement services

,J, Total length of case

,J, Total length of case

t Total length of case

t Length of time in out-ofhome placement in general

t Initial reporter law
enforcement or legal
professional, also school site
personnel
t Ass~nment to familY.
reuni 1cation and family
maintenance services
tinitial out-of-home placement
at Children's Shelter or foster
family home
t Mothers ordered to complete
parent education services

t Average stay in each out-ofhome placement
t Length of time in out-ofhome placement in general
(longest)

t

Case bein~ closed with
permanent p acement services

t Total leni!th of case

119

Overall Conclusions
Combining results from our overall sample of 1720 closed cases, the in-depth record
review of 403 child welfare cases and interviews with 8 key informants in managerial positions
in DFCS, the following overall conclusions can be drawn:

1.

Descriptive narratives presented in Section IX of this report indicate that the families
belonging to each ofthe four racial/ethnic groups present unique and diverse profiles.
•

African American families are largely headed by young, single, isolated, poor
mothers who suffer from substance abuse and have experienced criminal involvement
and domestic violence. At the same time, their reported rate of mental health
problems is low.

•

White families are made up of mothers who are also single, but who are better
educated than average and who are less likely to be on welfare. While substance
abuse, mental health problems and domestic violence are prevalent within this group,
white mothers tend to have a lower than average rate of involvement in the criminal
justice system.

•

Latino families are predominantly poor, with a higher than average number of
children. Many Latina mothers (18.4%) are non-English speaking and approximately
26% are foreign born. Latina mothers are more often living without a spouse, have
low levels of education and experience problems including substance abuse, and
criminal activity, as well as domestic violence. Like African American mothers, their
reported rates of mental health problems are low. For the immigrant Latino families,
issues related to immigration, legal status and the· ability to secure employment may .
also have an impact on their welfare, but were .undetectable with the available data.

•

Asian/Pacific Islander families who are assigned to involuntary child welfare
services, are also often non-English speaking and most likely to be foreign born.
Most Asian/Pacific Islander mothers in this group are older, have fewer children than
average, have less education than average and are married and living with a spouse.
Mental health problems and involvement in the mental health system rather than
substance abuse or criminality characterize this group. This finding is notable as
available research indicates that elevated rates of mental health problems are
uncharacteristic of immigrant populations. In addition, rates of physical abuse are
higher than average for this group and Asian American/Pacific Islander fathers are
comparatively more often reported as the perpetrators of abuse.

2. Families within each ofthe four racial/ethnic groups are extremely vulnerable.
•

The characteristics that emerged in our narrative descriptions as key characteristics of
each of the four racial/ethnic groups have been associated in the research literature as

120

risk factors for child abuse and neglect. These characteristics have also been
associated with increased serious behavioral and adjustment problems in children and
adolescents.
•

Comments by key informants underscored the relationship between characteristics
including poverty, lack of education, insufficient job skills, as well as involvement
with drugs and violence, and bias on the part of workers, as contributing to the
overrepresentation of African American and Latino children in the CWS.

•

While Asian American/Pacific Islander families have traditionally been
underrepresented in the CWS, our results suggest that Asian American/Pacific
Islander families with certain characteristics are emerging as a high-risk group. More
information and careful monitoring of this group is clearly warranted.

•

Identifying vulnerable families is an important piece of the puzzle in explaining
involvement and retention in the CWS. Understanding family characteristics that
represent risk factors can help systems design interventions that offset the potential
for CWS involvement and prolonged involvement.

3. Once in the Child Welfare System, children in each ofthe four racial/ethnic groups follow

different pathways and experience different outcomes.
•

African American children are younger than average when they enter the system, have
higher than average rates of being assigned to family reunification services and are
initially placed with a relative. African American children experience more court
hearings, have a higher than average rate of being removed from their families, longer
than average stays in each out-of-home placement and a longer average total case
duration than children in other groups. At case closure, African American children are
less likely to be reunified with their families and most frequently in permanent
placement.

•

White children are older than average on entry to the CWS and though they are also most
often assigned to family reunification services and placed with a relative, they tend to
have a shorter than average stay in each out of home placement and a lower than average
total case length. At case closure, White children are most likely to be in family
maintenance services.

•

Likewise, Latino children are also most often assigned to family reunification services
and placed with a relative at initial placement. They experience shorter than average
stays in each out-of-home placement, have a shorter total length of time in out-of-home
placement and a shorter total case length. Results also indicate that Latino children
experience a relatively high number of unique placement homes, suggesting that they
experience multiple transitions as they wend their way through the system. For Latino
children, their most common status at case closure is permanent placement, suggesting

121

that many Latino children are not reunified with their family of origin. Similar to African
American families, they experience a lower than average rate of family maintenance
services.
•

Asian American/Pacific Islander children also tend to be assigned to family reunification
services, but receive family maintenance services at a higher rate than other groups.
Rather than being placed with relatives, their initial placement is likely to be in a family
foster home. Asian American/Pacific Islander children appear to have lengthier than
average stays in each out-of-home placement and longer than average total time in outof-home placement, as well as a longer than average total case length. The most
frequently occurring final out-of-home placement for Asian American/Pacific Islander
children tends to be a family foster home, and they are more likely to have their case
closed with permanent placement services.

•

Once in the CWS, African American and Latino children tend not to return to their
families. This finding was underscored by focus group results in Phase I of our study, as
well as by comments from key informants during the current phase. Key informants
stated that worker bias, poverty, as well as immigration status were key barriers to exiting
the system. In addition, key informants suggested that individual bias on the part of
workers might be instrumental at both the front end of the system and in decisions that
are made once the child enters the system, in prolonging stays of children of color.

4. The services orderedfor families ofcolor are generally limited to a one-size-fits-all
approach and to a small array ofavailable services. The reliance on traditional formal
services does not appear to meet the needs of these highly diverse ethnic/racial family
groups.
•

Parenting education, substance abuse treatment and counseling (both individual and
group) were the most commonly ordered services at the jurisdictional/dispositional
hearing. These results suggest that workers are offering the same services for all
families, regardless of their needs and rely heavily on traditional, formal services.

•

Results also indicate that services are not distributed uniformly across racial/ethnic
groups and do not necessarily match their specific needs. In spite of high rates of
substance abuse problems and criminal involvement, African American parents were
most often ordered to receive parent education only. White families appear to receive
more of the available services including parenting education services, individual and
group counseling and substance abuse treatment services. In addition to parenting
education services, Latino families receive slightly higher than average rates of substance
abuse services. However, it is not clear whether these provided services are culturally and
linguistically appropriate. Neither African American nor Latino families appear to
receive mental health services. Asian American families appear to receive mental health
services at a higher rate than other groups, but again it is not clear whether they are
meeting the unique needs of this culturally distinct group.

122

•

The paucity of services ordered indicates that children and families of color are not
provided with sufficient preventive and supportive services and that traditional CWS may
not meet the particular needs of these unique and diverse groups. These findings also
corroborate those ofa national forum on children and families of color in the CWS
(CWLA, 2002). In addition, research has indicated that individual, group and couples
counseling, as well as parenting training have shown only minimal success in teaching
parents better skills and reducing the likelihood of further abuse in families marked by
serious and chronic abuse (Albee & Gullota, 1997; McLoyd, 1998).

•

There is a need for preventive and early intervention services for vulnerable families of
color. For example, home visitation services may be exceptionally helpful in addressing
the needs of vulnerable, at-risk families from diverse racial/ethnic groups. Research has
indicated that programs of home visitation that promote positive health-related behaviors
in mothers of young children, competent care of their children and linkage with needed
health care and human services, reduce rates of criminality, problems related to substance
abuse and child abuse and neglect among young, unmarried, isolated poor mothers (Olds
et al., 1997, 1998). Studies that follow children of mothers involved in such home
visitation programs, into adolescence have found these youth have fewer serious
behavioral and adjustment problems than youth whose mothers did not participate in
home visitation programs. Indeed, the U.S. Advisory Board on Child Abuse and Neglect
has recommended that home-visitation services be made available to all parents of young
children as a means of preventing child abuse and neglect (U.S. Department of Health
and Human Services, Administration for Children and Families, 1991).

•

The use of more non-traditional, culturally sensitive services that are conducted in the
client's primary language are clearly -warranted. Key informants underscored this point.
A paucity of social services, particularly multi-lingual services, was cited as a significant
barrier for many families of color. Interviewees discussed the shortage of substance abuse
treatment programs, particularly those geared for women with children and people whose
primary language is not English. Multi-lingual and culturally appropriate domestic
violence services, parenting classes, and other social services were considered in need of
development.

•

The form of mandated services is also an important issue. As one key informant noted,
"if a Latino parent is court ordered to therapy, this may be seen as 'being crazy, while
attending a psychoeducational group is (viewed as) less of a problem."

•

The scarcity of accessible services was perceived as "discrimination against the poor who
have to rely on free treatment" by more than one informant.

•

In addition, key informants noted that recognition of the extended family system is
critical, particularly in relation to working with many communities of color. "Thinking of
family as a mother-father-child configuration is common but problematic .. .it is just not
how children are really raised."

123

•

More research and evaluation of interventions and programs targeted to children and
families of color is necessary. Rather then implementing untested interventions, it would
be more efficacious to begin with programs that have been tested, replicated and found to
work with families from unique and diverse racial/ethnic groups.

5. There is a need to involve multiple social service systems in a comprehensive and
coordinated effort to meet the needs ofchildren and families ofcolor.
•

Results from Phase 2 indicate that the problems experienced by families across the
different racial/ethnic groups span multiple systems including: mental health, juvenile
justice, adult criminal justice, substance abuse, and welfare.

•

Statistics presented in the literature review section of this report also indicate that
families of color are involved in systems other than child welfare in high numbers.
Prevention and intervention efforts should involve a deliberate and organized
coordination of these multiple systems.

124

References
Albee, G.W. & Gullota, T.P. (Eds.) (1997). Primary Prevention Works. Thousand Oaks:
Sage Publications.
Appel, A.E., & Holden, G.W. (1998). The co-occurrence of spouse and physical child
abuse: A review and appraisal. Journal of Family Psychology, 12, 578-599.
Ards, S. (1992). Understanding patterns of child maltreatment. Contemporary Policy
Issues, 10, 39-50.
Ards, S., & Harrell, A. ( 1993). Reporting of child maltreatment: A secondary analysis of
the national incidence surveys. Child Abuse and Neglect. 17. 337-344.
Ards, S., Chung, C., & Myers Jr .. , S., L. (1998). The effects of sample selection bias on
racial differences in child abuse reporting. Child Abuse and Neglect. 22. 103-116.
Bays, J. (1990). Substance abuse and child neglect: Impact of addiction on the child.
Pediatric Clinics of North America, 3 7. 881-904.
Beck, A., Gilliard, D., Greenfeld, L., Harlow, C., Hester, T., Janowski, L., Snell, T.,
Stephen, J., & Morton, D. (1993). Survey of state prison inmates, 1991. Washington DC: U.S.
Department of Justices Bureau of Justice Statistics.
Beck, A. J., & Harrison, P. M. (2001). Prisoners in 2000. Washington DC: U.S.
Department of Justice, Bureau of Justice Statistics.
Beck, A. J. & Karberg, J.C. (2001). Prison and Jail Inmates at Midyear 2000.
Washington DC: U.S. Department of Justice, Bureau of Justice Statistics.
Bellis, M. D., Broussard, E. R., Herring, D. J., Wexler, S., Moritz, G., & Benitez, J. G.
(2001). Psychiatric co-morbidity in caregivers and children involved in maltreatment: A pilot
research study with policy implications. Child Abuse & Neglect 25. 923-944.
Bowen, K. (2000). Child abuse and domestic violence in families of children seen for
suspected sexual abuse. Clinical Pediatrics, 39. 33-40.
Brooks-Gunn, J., Duncan, G. J., & Maritato, N. (1997). Poor families, poor outcomes:
The well-being of children and youth. In G. J. Duncan & J. Brooks-Gunn (Eds.), Consequences
of growing up poor (pp. 1-17). New York: Russell Sage.
Burstein &Woodsmall (1999). 1999 Santa Clara County Homeless Survey. San Jose
State University Community Outreach Partnership Center.

125

Chaffin, M., Kelleher, K., & Hollenberg, J. (1996). Onset of physical abuse and neglect:
Psychiatric, substance abuse and social risk factors from prospective community data. Child
abuse and neglect, 20, 191-203.
Child Welfare League of America (1998). State survey on children with incarcerated
parents. [Online]. Available:
http://www.cwla.org/programs/incarcerated/98stateagencysurvey.htm.
Child Welfare League of America (2000). The impact of AFSA on families and children
of color: Proceedings of a forum, November, 2000. Washington, DC: CWLA, Inc.
City of San Jose, Department of Parks, Recreation and Neighborhood Services, Youth
Services Division (2001). Youth Services Master Plan: A Blueprint for Bridging the Digital
Divide. San Jose, CA: Author.
Coulton, C. J., Korbin, J.E., & Su, M. (1999). Neighborhoods and child maltreatment: A
multi-level study. Child Abuse and Neglect, 23, 1019-1040.
Coulton, C., I., Korbin, I.E., Su, M., & Chow, I. (1995). Community level factors and
child maltreatment rates. Child Development. 66, 1262-1276.
Coulton, C., & Pandey, S. (1992). Geographic concentration of poverty and risk to
children in urban neighborhoods. American Behavioral Scientist, 35, 238-257.
Courtney, M.E., Barth, R.P., Duerr Berrick, J., Brooks, D., Needell, B., & Park, L.
(1996). Race and child welfare services: Past research and future directions. Child Welfare, 75,
99-137.
Dinwiddie, S. H., & Bucholz, K. K. (1993). Psychiatric diagnoses of self-reported child
abuseers. Child Abuse & Neglect, 17, 465-476.
Drake, B., & Pandey, S. {1996). Understanding the relationship between neighborhood
poverty and specific types of child maltreatment. Child Abuse and Neglect, 20, 1003-1018.
Duncan, G. J. (1991). The economic environment of childhood. In A. C. Huston (Ed.),
Children in poverty: Child development and public policy (pp. 23-50). Cambridge, MA:
Cambridge University.
Egami, Y., Ford, D. E., Greenfield, S. F., & Crum, R. M. (1996). Psychiatric profile and
sociodemographic characteristics of adults who report physically abusing or neglecting children .
. American Journal of Psychiatry, 153, 921-929.
Enchautegui, M. E. (1997). Latino neighborhoods and Latino neighborhood poverty.

126

Journal of Urban Affairs, 19, 445-467.
Famularo, R., Kinscherff, R., & Fenton, T. (1992). Psychiatric diagnoses of abusive
mothers. The Journal ofNervous and Mental Disease, 180, 658-661.
Famularo, R., Barnum, R., & Stone, K. ( 1986). Court-ordered removal in severe child
maltreatment: An association to parental major affective disorder. Child Abuse and Neglect, 10,
487-492.
Federal Interagency Forum on Child and Family Statistics. America's Children: Key
National Indicators of Well-being, 2001. Federal lnteragency Forum on Child and Family
Statistics, Washington DC: U.S. Government Printing Office.
Garbarino, J., & Sherman, D. (1980). High-risk neighborhoods and high-risk families:
· The human ecology of child maltreatment. Child Development, 51. 188-189.
Gentry, P.M. (1998). Permanency planning in the context of parental incarceration: Legal
issues and recommendations. Child Welfare. 77, 543-559.
Greenfeld, L.A., and Snell, T. L. (1999). Women Offenders. Washington DC: U.S.
Department of Justice, Bureau of Justice Statistics.
Gilliard, D., & Beck, A. (1998). Prisoners in 1997. Washington DC: U.S. Department of
Justice, Bureau of Justice Statistics.
Hampton, R. L., & Newberger, E. H, (1985). Child abuse incidence and reporting by
hospitals: Significance of severity, class and race. American Journal of Public Health, 75, 56-60.
Hines, A.M., Merdinger, J. Lee, P.A. & Coach, M. (2001). An Evaluation of Factors
Related to the Disproportionate Representation of Children of Color in Santa Clara County's
Child Welfare System. Report submitted to Santa Clara County, CA Department of Social
Services, Children and Family Division, March 30, 2001.
Hollinger, J. H. (1998). A Guide to the Multi-Ethnic Placement Act of 1994 as amended
by the Interethnic Adoption Provisions of 1996. Washington, DC: American Bar Association.
[Online] Available: http://www.acf.dhhs.gov/programs/cb/publications/mepa94/mepachp3 .htm.
Huston, A. C. (1991). Children in poverty: Development and policy issues. In A. C.
Huston (Ed.), Children in poverty: Child development .and public policy (pp. 1-22). Cambridge,
MA: Cambridge University.
Jaudes, P. K., Ekwo, E., & Voorhis, J. V. ( 1995). Association of drug abuse and child
abuse. Child Abuse and Neglect, 19, 1065-1076.

127

Joint Venture: Silicon Valley Network (2001 and 2002). Index of Silicon Valley:
Measuring Progress toward Silicon Valley 2010. San Jose CA: Author.
Katz, L. (1998). Concurrent Planning: A trainer's guide. Seattle, WA: Northwest Institute
for Children and Families.
Kids in Common (2000). Silicon Valley Children's Report Card. San Jose, CA: Author.
Knapp, S.A., McDonald, T.P., & Diamond, K.L. (2001). The path to adoption for
children of color. Child Abuse and Neglect, 25, 215-229.
Kotch, J.B., Browne, D. C., Dufort, V., Winsor, J., & Catellier, D. (1999). Predicting
child maltreatment in the first 4 years of life from characteristics assessed in the neonatal period.
Child Abuse & Neglect, 23, 305-319.
Lightbourne, W. (2001). Quarterly Statistical Data of Public Assistance Families in the
County of Santa Clara County by Cities. County of Santa Clara, CA: Department of
Employment and Benefits Data Services.
Massey, D.S., & Denton, N. A. (1989). Hypersegregation in U.S. metropolitan areas:
Black and Hispanic segregation along five dimensions. Demography. 26, 373-391.
Massey, D., & Denton, N. A. (1993). American apartheid: Segregation and the making of
the underclass. Cambridge, MA: Harvard University Press.
McGuigan, W., M., & Pratt, C. C. (2001 ). The predictive impact of domestic violence on
three types of child maltreatment. Child Abuse & Neglect, 25. 869-883.
McLoyd, V.C. (1998). Socioeconomic disadvantage and child development. American
Psychologist, 53, 185-204.
McLoyd, V.C., & Wilson, L. (1991). The strain of living poor: Parenting, social support,
and child mental health. In A. C. Huston (Ed.), Children in poverty: Child development and
public policy (pp. 105-135). New York: Cambridge University Press.
Miranda, L. C. (1991). Latino child ·poverty in the United States. Washington, D.C.:
Children's Defense Fund.
Moore, K., & Vandivere, S. (2000). Children's Family Environment. Snapshots of
American Families II: A View of the Nation and 13 States from the National Survey of
American Families. Washington, D.C.: Urban Institute.
Mumola, C.J. (2000). Special Report: Incarcerated parents and their children.

128

Washington DC: U.S. Department of Justice, Bureau of Justice Statistics.
Naifeh, M. ( 1998). Dynamics of economic well-being, poverty 1993-1994: Trap door?
Revolving door? Or both? [Online] U.S. Census Bureau. Available:
http://www.census.gov/ftp/pub/hhes/poverty/povdynam/pov9394.html
Needell, B., Webster, D., Cuccaro-Alamin, S., Armijo, M., Lee, S., Brookhart, A., Lery,
B., Shaw, T., & Kim, H. (2002). Performance Indicators. Retrieved [8/16/02] from University of
California at Berkeley Center for Social Services Research website, URL
<http://cssr.berkeley.edu/CWSCMSreports/>.
Olds, D. et al., (1997). Long term effects of home visitation on maternal life course and
child abuse and neglect: fifteen year follow-up of a randomized trial. JAMA, 8.637-643.
Olds, D., et al (1998). Long term effects of nurse home visitation on children's criminal
and antisocial behavior. JAMA, 280. 1238~1244.
Ortega, R. M., Guillen, C., & Gutierrez-Najera, L. (1996). Latinos and child
welfare/Latinos y el bienestar del Nino: Voces de la comunidad. Ann Arbor, MI: The National
Latino Child Welfare Advocacy Group.
Pecora, P. J., Fraser, M.W., Nelson, K.E., Mccroskey, J., & Meezan, W. (1995).
Evaluating family-based services. New York: Airline de Gruyter.
Pelton, L. (1978). Child abuse and neglect: the myth of classlessness. American Journal
of Orthopsychiatry, 48, 608-617.
Public Policy Institute of California (2001). California Population. [Online] Available:
http://www.ppic.org/
Santa Clara County Children and Families First Commission (2000). A Chance for Evei:y
Child: Proposition 10 Strategic Plan. Santa Clara County, California: Author.
Santa Clara County Department of Corrections (2001). Annual Report. [Online].
Available: http://www.santaclaradoc.org/.
Santa Clara County Department of Employment and Benefit Services (2001). CalWorks
Plan. Santa Clara County, CA: Author.
Santa Clara Valley Health and Hospital System (2001). Annual Data Report 2001:
Domestic Violence in Santa Clara County. [Online]. Available:
http://www.sccphd.org/statistics2/Reports/.
Santa Clara Valley Health and Hospital System ( 1997). Health Status Report 1997.

129

[Online]. Available: http://www.sccphd.org/statistics2/Reports/.
Saunders, E. J., Nelson, K., & Landsman, M. J. (1993). Racial inequality and child
neglect: Findings in a metropolitan area. Child Welfare, 4, 341-354.
Schwartz, I.M. & Fishman, G. (1999). Kids raised by the government.Westport: Praeger.
Snell, T. (1994). Special Report: Women in prison. Washington DC: U.S. Department of
Justice, Bureau of Justice Statistics.
Sedlak, A., J., & Broadhurst, D. D. (1996). Third National Incidence Study of Child
Abuse and Neglect. Washington, D.C.: U.S. Department of Health and Human Services,
National Center on Child Abuse and Neglect.
Semiedi, J., Radel, L.F., & Nolan, C. (2001). Substance abuse and child welfare: Clear
linkages and promising responses. Child Welfare, 80, 109-128.
State of California, Department of Alcohol and Drug Program (1998) CADDS . Cited in
Santa Clara County Public Health Department (1999). Key Indicators of Well-being 1999. Santa
Clara County CA: Author.
Staveteig, S., & Wigton, A. (2000). Racial and Ethnic Disparities: Key Findings from the
National Survey of American Families. Washington, D.C.: Urban Institute.
Steinberg, L.D., Catalano, R., & Dooley, D. (1981). Economic antecedents of child
abuse and neglect. Child Development, 52(3), 975-985.
Tajima, E. A. (2000). The relative importance of wife abuse as a risk factor for violence
against children. Child Abuse & Neglect, 24, 1383-1398.
United States Census Bureau (1999). Povercy: 1998 highlights from the Current
Population Survey. [Online] Available:
http://www.census.gov/ftp/pub/hhes/poverty/poverty98/pov98hi.html
United States Census Bureau (2001). Overview of Race and Hispanic Origin, Census
2000 Brief. [Online] Available: http://www.census.gov/prod/200lpubs/c2kbr0l-l.pdf
United States Census Bureau (2001). Santa Clara County Quick Facts 2000. [Online]
Available: http://quickfacts.census.gov/qfd/states/06000.html.
United States Department of Health and Human Services, Children's Bureau. (1997).
National study of protective, preventive and reunification services delivered to children and their
families. Washington DC: U.S. Government Printing Office.

130

United States Department of Health and Human Services' National Center on Child
Abuse and Neglect (1998). Child maltreatment 1997: Reports from the states to the national
center on child abuse and neglect. Washington, D.C.: Author.
United States Department of Health and Human Services (1999). Adoption 2002: The
President's initiative on adoption and foster care. Guidelines for public policy and state
legislation governing permanence for children. Washington, D.C.: Author.
United States Department of Health and Human Services (1999). Blending perspectives
and building common ground: A report to Congress on substance abuse and child protection.
Washington, D.C.: Author.
United State General Accounting Office (1999). Foster care: States' early experiences
implementing the Adoption and Safe Families Act. HEHS-00-1. Washington, D.C.: Government
Printing Office.
United State Government Accounting Office (1998). Foster Care: Agencies face
challenges securing stable homes for children of substance abusers. HEHS-98-182. Washington,
D.C.: Government Printing Office.
Vega, W., Kolody, B., Hwang, J., & Noble, A. (1993). Prevalence and magnitude of
perinatal substance exposures in California. New England Journal of Medicine, 329, 850-854.
Walker, C., Zangrillo, P. & Smith, J. (1991). Parental drug abuse and African American
children in foster care: Issues and study findings (Publication No. SA-90-2233-1). Washington
D.C.: U.S. Department of Health and Human Services, Office of the Assistant Secretary for
Planning and Evaluation, Division of Children and Youth Policy.
Weil, A. & Feingold, K. (2002). Welfare Reform: The Next Act. Washington, DC: The
Urban Institute Press.
Zambrana, R. E. & Dorrington, C. (1998). Economic and social vulnerability of Latino
children and families by subgroup: Implications for child welfare. Child Welfare, 77, 5-27.
Zedlewski, S. (2000). Family Economic Well-Being. Snapshots of American Families II:
A View of the Nation and 13 States from the National Survey of American Families.
Washington, D.C.: Urban Institute.
Zellman, G. (1990a). Report decision-making patterns among mandated child abuse
reporters. Child Abuse and Neglect. 14, 325-336.
Zellman, G. (1990b). Child abuse reporting and failure to report among mandated
reporters: Prevalence, incidence and reasons. Journal ofintemersonal Violence, 5, 3-22.

131

Zellman, G. (1992). The impact of case characteristics on child abuse reporting decisions.
Child Abuse and Neglect, 16, 57-74.
Zuravin, S. J. (1989). Severity of maternal depression and three types of mother-to-child
aggression. American Journal of Orthopsychiatry, 59, 377-390.

ATTACHMENT 1:

Sampling Procedures

Sampling Procedures
We sought a representative sample from the 1753 closed cases making up the
population of concern. In achieving representativeness, we were mindful of a need to
include a sufficient numbers of children and youth in our sample of certain types to
permit us to address powerfully, from a statistical point of view, questions central to the
study. Factors where we considered it important to guarantee the presence of large
numbers of children and youth were ethnicity, age, and service type.

Stratification. To insure that enough children and youth were included from each
minority group, across the age spectrum, and from each service type, we adopted
stratified random sampling. We grouped the 1753 closed cases on the basis of child
ethnicity (Black, White, Latino, Asian), age, (0-6, 7-12, >13), and service type
(emergency response vs. family maintenance vs. family reunification vs. permanent
placement).
Regarding categories of service type, we wished to accurately reflect the fact that
among the closed cases, more children would appear in permanent placement and family
reunification than in family maintenance and that concern was great to understand the
experience of children in all groups. Accordingly, we initially devised our sampling
strategy such that approximately 25% of children in the sample would come from the
family maintenance group.
Ultimately, our choice of stratification variables reflected a trade-off often faced
by researchers as they devise sampling plans. In choosing strata (e.g. age) and groups
within strata (e.g. 0-6, 7-12, 13 or more), we sought a balance between preserving
important distinctions on the one hand, and creating a practical and workable scheme on

the other. The resulting system of forty-eight individual categories (four ethnic groups x
three age groups x four service categories) was considered to reflect as many strata and
categories as were feasible to include under the circumstances.
We understood also that, from subsequent dialogue with advisors to the study,
groups of special concern might be identified later. If such groups were underrepresented,
the final sample could be supplemented readily by random selection or, if necessary, by
including all available group members.
Sample size. In keeping with the original plan for the study, our goal was to have

400 records for our final sample. We recognized, however, that not all closed cases
drawn for the sample would be available for coding to case record reviewers when they
sought access. We therefore targeted 500 cases for inclusions in our sample. From 500
cases, our final sample would include 400 cases or more if up to 100 cases were not
available when access to records ultimately was sought.
Projecting a sample size of 500, we allocated a specific number of cases-that is,
a quota-- for selection from each of the forty eight categories formed by combing
ethnicity, age, and service type as described above. We then randomly chose from each
category a number of cases matching the quota. For each category, if fewer cases were
available than the quota, we took all cases in the group. The shortfall in cases-the
number of cases that were not available in a category with too few cases to meet the
quote-was reassigned proportionally to other categories.
Following this stratified random sampling procedure, we chose 504 cases for our
sample. There were 4.4% (n=22) emergency response cases, 37.7% (n=190) family
maintenance, 8.5% (n=43) family reunification, and 49.4% (n=249) permanent

placement. Considering age, there were 44.4% (n=224) ages 0-6, 31 % (n=l56) ages 7-12,
and 124 (n=24.6%) ages 13 and above. The ethnic composition of the sample was 19.4%
(n=98) African American, 31.9% (n= 161) Latino, 14.9% (n=75) Asian/Pacific Islander,
and 27.8% (n=l40) white.
Supplemental sampling. Initial attempts to draw cases for our sample indicated
that initial estimates of the number of missing cases and cases in Voluntary Family
Maintenance were too low. Fearing having fewer than 400 cases, the desired, final
sample size, we chose to supplement our sample with 200 additional cases. We randomly
chose 200 cases from our previously constructed strata, following procedures outlined
above. These cases will be added ifwe exhaust our initial selection of 504 cases before
meeting our target of a final sample of size 400.

ATTACHMENT 2:
Data Collection Process
and Procedures

Data Collection Process and Procedures

Case record reviews are currently taking place at the Social Services Record
Retention Center. Ten masters-level social work students are each reviewing four.cases a
week. Kathy Lemon MSW, research associate, is supervising students and overseeing the
data collection process. She is also working with the retention center manager, Maria
Gonzalez to insure that 40 cases are pulled and set aside in a designated area for students
to review each week. Cases are pulled from the main sample list of 400 cases~ Students
have been randomly assigned to review 40 cases each.
If a case if missing or not on the shelf, retention center staff move.to the next case
on the list. A separate list of the missing cases has been generated and retention center
staff will put out inquiries regarding these cases to see if they can be called back to the
retention center to be reviewed. If a case cannot be called back, we are working to find
out the reasons why a cases is missing in order to insure that missing cases do not bias the
overall sample. Maria and all of the staff at the retention center have been extremely
helpful in pulling cases on an as needed basis and accommodating the students who are
reviewing cases at the retention center.
Once cases have been pulled, Kathy then inventories the cases and verifies that
they are a part of the sample and removes any voluntary family maintenance cases.
Voluntary family maintenance cases are being removed from the main sample because
they do not contain the court records from which students are gathering data. As such, in
order to maintain the reliability of the data collection, these cases are being pulled from

the main sample and kept in a designated area. A separate analysis of these cases is being
considered.
Students review cases at the retention center. When they are finished with a case,
they leave the case in box to be re-shelved and they leave the data extraction form in a
designated box for Kathy to pick up. Kathy then inventories the cases that are completed
and keeps track of the students' progress in data collection.

ATTACHMENT 3:
Child Welfare System Case Record Data
Extraction Form

*** CONFIDENTIAL INFORMA T/ON ***
Santa Clara County / SJSU College of Social Work
CWS Case Record Data Extraction Form

Child Characteristics (at time of incident)

SPSS

Item#
C1.

Child Characteristics (at time of incident)
Sex
Female

C2.

Ethnicity

C2a.

Speaks English?

NIA - too young to speak
Not Indicated

c.eng

C2b.

Translator needed? (only if non-English
speaking} Indicate language

Not Indicated
NIA - too young to speak

c.plang

c.sex
c.eth

Language
C3.

Type of abuse/neglect,
Sustained by court in minute order
(W&I 300 codes a-j}

C4.

Number of prior referrals

cs.

a - Serious physical harm
b - General neglect
c - Emotional abuse
d - Sexual abuse
e - Serious physical harm of child < 5
f - Death of sibling through
abuse/neglect by parent or guardian
g - Caretaker absence/incapacity
h - Relinquished
i - Cruelty
j - Child's sibling has been abused or
neglected as defined in a, b, d, or i

c.abng

c.pr#ref

Previous # times in CWS
(recipient of CWS services}

c.pr#cws

C6.

Initial Reporter of Incident (police
report: reporting party, redacted}

c.reptr

C6a.

Date of incident

C7.

Date of birth

cs.

Place of Birth (city, state, country}

Not Indicated

C9.

Born in US?

Not Indicated

C10.

IF NOT BORN IN US - Length of time
in US (specify# davs, months or years)
Length of time in California
(specify# days, months or years)

Not Indicated

C11.

c.datei
c.bday

Not Indicated

c.pob

c;brnusa
c.tmeusa
c.tmeca

Item#
C12.

SPSS

Child Characteristics (at time of incident)
Last grade completed in school
Indicate Grade:
Not old enough for school
Not Indicated

C13.

C13a.
C14.
C14a.
C15.
C15a.
C15b.

C16.
C16a.
C17.
C17a.
C18.
C18a.

Academic problems (including truancy
and dropped out)
If yes, specify type
Substance Abuse?

Disability?
{physical, learning, etc.)
If yes, specify type
If yes, what assistance was
provided? (e.g., signer for hearing
impaired)
Developmental delays?

Emotional problems?

If yes, specify

C21

Not Indicated

Gender identity issue?

No Assistance
Not Indicated
Not Indicated

c.disas

c.dvdl
c.dvdlt

Not Indicated

c.ment
c.mentt

Not Indicated

c.emot
c.emott

Not Indicated

c.bhv
c.bhvt

Not Indicated

If yes, specify
Criminal History (outstanding warrants,
arrests, or convictions)?

c.dis
c.dist

If yes, specify

C19a.

c.sbhx
c.sbhxt

If yes, specify diagnosis

Behavior problems?

C20a.

Not Indicated

If yes, specify type
Mental Illness?

c.acpb

c.acpbt

If yes, what substance

C19.

C20.

Not Indicated

c.educ

c.geni
c.genit

Not Indicated

c.crhx

Item#

C21a.

C22.

Child Characteristics (at time of incident)

SPSS

If yes, specify crime(s)

Current health status

c.crhxt

Healthy / No health problems

c.hths

Excellent
Fair
Poor
Other evaluation
Not Indicated
C23.
C23a.

Item#

C24.

Participation in other service systems?
If yes, specify ALL service systems

Child Characteristics (at time of incident)

Participation in other services?

Not Indicated
Mental Health
Health
Juvenile Justice
Other {specify)

c.ptsv
c.ptsvt

Not Indicated

SPSS
c.posv

C24a.

C25.

If yes, specify ALL services, (e.g.
child advocacy, mentorship,
counseling ... )

Child's current living situation
(check all that apply)

c.posvt

alone
with mother
with father
with both mother and father
with alternate caretaker
specify who
with other(s), specify person(s):

not indicated

c.crls

Mother Characteristics (at time of incident)
Item#
M1.

Mother Characteristics (at time of incident)
Ethnicity

M2.

Date of Birth

M3.

Speaks English?

Not Indicated

M4.

Translator needed? {only if non-English
speaking) Indicate language

Not Indicated

SPSS
m.eth
m.bday

Language

M5.

Place of Birth {city, state, country)

Not Indicated

MS.

Born in US?

Not Indicated

M7.

IF NOT BORN IN US - Length of time
in US {specify# days, months or years)

Not Indicated

MB.

Length of time in California
{specify# days, months or years)

Not Indicated

M9.

Currently Employed?

Not Indicated

M10.

Last Occupation

Not Indicated

M11.

Last grade completed in school

Not Indicated

M12.

Marital Status

M13.

Not Indicated

M14.

Family Household Income
(state as specifically as possible)
Welfare Eligible?

M15.

Substance Abuse?

Not Indicated

M15a.
M16.
M16a.
M16b.

M17.

Married
Remarried, if so what#
Single, never married
Separated
Divorced
Widowed
Other
Not Indicated

If yes, what assistance was
provided? {e.g., signer for hearing
impaired)
Developmental delays?

m.plang

m.pob
m.brnusa
m.tmeusa

m.tmeca

m.cremp
m.lstjob
m.educ
m.mstat

Not Indicated

If yes, what substance
Disability?
(ohvsical, learning, etc.)
If yes, specify type

m.eng

m.inc
m.wfe
m.sbhx
m.sbhxt

Not Indicated

m.dis
m.dist

No Assistance
Not Indicated
Not Indicated

m.disas

m.dvdl

Item#
M17a.

Mother Characteristics (at time of incident)
If yes, specify type

M18.

Mental Illness?

M18a.

M19.
M19a.

Emotional problems?

M21a.

If yes, specify
Gender identity issue?

M23a.

Item#
M24.

Currently Incarcerated?

Mother Characteristics (at time of incident)

m.ment
m.mentt

Not Indicated

m.emot
m.emott

Not Indicated

m.bhv
m.bhvt

Not Indicated

If yes, specify
Criminal History (outstanding warrants,
arrests, or convictions)?
If yes, specify crime(s)

M23.

Not Indicated

If yes, specify
Behavior problems?

M22a.

m.dvdlt

If yes, specify diagnosis

M20.

M22.

SPSS

m.geni
m.genit

Not Indicated

m.crhx
m.crhxt

Not Indicated

SPSS
m.carc

M24a.
M25.

If yes, specify location
Current health status

Healthy / No health problems
Excellent

m.carcl
m.hths

Fair
Poor
Other evaluation
Not Indicated
M26.
M26a.

M27.
M27a.

Item#

Participation in other service systems?
If yes, specify ALL service systems

Not Indicated
Mental Health
Health
Criminal Justice
Other (specify)

Participation in other services?
If yes, specify ALL services, (e.g.
Counseling, anger management,
Parenting class ... )

Mother Characteristics (at time of incident)

m.ptsv
m.ptsvt

Not Indicated

m.posv
m.posvt

SPSS

M28.

Mother's current living situation
(check all that apply)

D alone
with spouse
with children (specify how many _ _ )
with extended family
with other(s), specify person(s):

m.crls

not indicated
M29.

Is mother the child's primary caretaker?

Not Indicated

M30.

Does mother currently (do we mean at
time of referral?) have legal custody of
child?

Not Indicated

m.prctk
m.cust

Father Characteristics (at time of incident)
Item#
F1.

Father Characteristics Cat time of incident)
Ethnicity

F2.

Date of Birth

F3.

Speaks English?

Not Indicated

F4.

Translator needed? {only if non-English
speaking) Indicate language

Not Indicated

SPSS
f.eth
f.bday

Language

F5.

Place of Birth {city, state, country)

Not Indicated

F6.

Born in US?

Not Indicated

F7.

IF NOT BORN IN US - Length of time
in US {specify# days, months or years)

Not Indicated

F8.

Length of time in California
{specify# days, months or years)

Not Indicated

F9.

Currently Employed?

Not Indicated

F10.

Last Occupation

Not Indicated

F11.

Last grade completed in school

Not Indicated

F12.

Marital Status

.

Married
Remarried, if so what #
Single, never married
Separated
Divorced
Widowed
Other
Not Indicated

F14.

Not Indicated

F15.

Substance Abuse?

Not Indicated

F15a.
F16.
F16a.
F16b.

If yes, what assistance was
provided? (e.g., signer for hearing
impaired}

f.pob
f.bmusa
f.tmeusa

f.tmeca

f.cremp
f.lstjob
f.educ

f.inc

If yes, what substance
Disability?
(physical, learnina, etc.)
If yes, specify type

f.plang

f.mstat

Family Household Income
(state as specifically as possible)
Welfare Eligible?

F13.

f.eng

f.wfe
f.sbhx
f.sbhxt

Not Indicated

f.dis
f.dist

No Assistance
Not Indicated

f.disas

Item#
F17.
F17a.
F18.
F18a.
F19.
F19a.

Father Characteristics (at time of incident)

Developmental delays?
If yes, specify type
Mental Illness?

Emotional problems?

F21a.

If yes, specify

F22a.
F23.
F23a.

Item#
F24.

Gender identity issue?

Currently Incarcerated?

f.emot

f.bhv
f.bhvt

Not Indicated

f.geni
f.genit

Not Indicated

If yes, specify crime{s)

Father Characteristics (at time of incident)

f.ment

f.emott

Not Indicated

If yes, specify
Criminal History {arrest or conviction)?

f.dvdl

f.mentt

Not Indicated

If yes, specify
Behavior problems?

SPSS

f.dvdlt

Not Indicated

If yes, specify diagnosis

F20.

F22.

Not Indicated

f.crhx
f.crhxt

Not Indicated

SPSS
f.carc

F24a.
F25.

If yes, specify location
Current health status

f.carcl

Healthy / No health problems
Excellent

f.hths

Fair
Poor
Other evaluation
Not Indicated
F26.
F26a.

F27.
F27a.

Item#

Participation in other service systems?
If yes, specify ALL service systems

Not Indicated
Mental Health
Health
D Criminal Justice
D Other (specify)

Participation in other services?
If yes, specify ALL services, (e.g.
Counseling, anger management,
Parenting class ... )

Father Characteristics (at time of incident)

f.ptsv
f.ptsvt

Not Indicated

f.posv
f.posvt

SPSS

F28.

Father's current living situation

D alone
with spouse
with children (specify how many _ _ )
D with extended family
with other(s}, specify person(s):

f.crls

not indicated
F29.

Is father the child's primary caretaker?

Not Indicated

F30.

Does father currently (do we mean at
time of referral?) have legal custody of
child?

Not Indicated

f.prctk
f.cust

Alternate Caretaker Characteristics (at time of incident)
(Assess if neither mother or father is the primary caretaker)
Item#
A1.
A2.

SPSS

Alternate Caretaker Characteristics (at time of incident)

Relationship to child

a.rel2ch

· Length of time as caretaker
(specify# days, months or years)

a.tmasct

A3.

Sex

A4.

Ethnicity

AS.

Date of Birth

A6.

Speaks English?

Not Indicated

A7.

Translator needed? (only if non-English
speaking) Indicate language

Not Indicated

Female

a.sex
a.eth
a.bday

Language

AB.

Place of Birth (city, state, country)

Not Indicated

'A9.

Born in US?

Not Indicated

A10.

IN NOT BORN IN US - Length of time
in US (specify# days, months or years)

A11.

Length of time in California
(specify# days, months or years)

Not Indicated

A12.

Currently Employed?

Not Indicated

A13.

Last Occupation

Not Indicated

A14.

Last grade completed in school

Not Indicated

A15.

Marital Status

A16.

Family Household Income
(state as specifically as possible)

Not Indicated

A17.

Welfare Eligible?

Not Indicated

A18.

Substance Abuse?

Not Indicated

A18a.

If yes, what substance

D Not Indicated

Married
Remarried, if so what #
Single, never married
Separated
Divorced
Widowed
Other
Not Indicated

a.eng
a.plang

a.pob
a.brnusa
a.tmeusa
a.tmeca
a.cremp
a.lstjob
a.educ
a.mstat

a.inc
a.wfe
a.sbhx
a.sbhxt

Item#
A19.
A19a.
A19b.

A20.
A20a.
A21.
A21a.

A22.
A22a.

Alternate Caretaker Characteristics (at time of incident)
Disability?
(physical, learning, etc.)
If yes, specify type
If yes, what assistance was
provided? {e.g., signer for hearing
impaired)
Developmental delays?

Mental Illness?

A23a.

If yes, specify

A25.
A25a.

Item#
A26.

Not Indicated

Gender identity issue?

Not Indicated

a.ment

a.emot
a.emott

Not Indicated

a.bhv
a.bhvt

Not Indicated

a.geni
a.genit

Not Indicated

If yes, specify crime{s)

Alternate Caretaker Characteristics (at time of incident)
Currently Incarcerated?

a.dvdl

a.mentt

Not Indicated

If yes, specify
Criminal History {outstanding warrants,
arrests, or conviction)?

a.disas

a.dvdlt

If yes, specify
Behavior problems?

A24a.

No Assistance
D Not Indicated

If yes, specify diagnosis
Emotional problems?

a.dis
a.dist

If yes, specify type

A23.

A24.

Not Indicated

SPSS

a.crhx

a.crhxt

Not Indicated

SPSS
a.care

A26a.
A27.

If yes, specify location
Current health status

Healthy / No health problems
Excellent

a.carcl
a.hths

Fair
Poor
Other evaluation
Not Indicated
A28.

A28a.

A29.

A29a.

Participation in other service systems?
If yesl specify ALL service systems

Not Indicated
Mental Health
Health
Criminal Justice
Other (specify)

Participation in other services?

a.ptsvt

Not Indicated

If yes, specify ALL services, (e.g.
Counseling anger management,
Parenting class ... )

Alternate caretaker1s current living
situation

a.posv
a.posvt

1

A30.

a.ptsv

alone
with spouse
with children (specify how many __ )
with extended family
with other(s), specify person(s):

a.eris

not indicated
A31.

Is alternate caretaker the child's
primary caretaker?

Not Indicated

a.prctk

Perpetrator Information (at time of incident)
Item#
P1.

Perpetrator Characteristics (at time of incident)
Perpetrator {relationship to child)

P2.

Is the perpetrator someone other than
the mother, father, or alternate
caretaker?

SPSS

p.rel2ch

No

If YES, finish section, if NO, skip
perpetrator information and
go to page 11.

P3.

Sex

P4.

Ethnicity

PS.

Date of Birth

P6.

Speaks English?

Not Indicated

P7.

Translator needed? {only if non-English
speaking) Indicate l~nguage

Not Indicated

Female

p.sex
p.eth
p.bday

Language

PB.

Place of Birth (city, state, country)

Not Indicated

P9.

Born in US?

Not Indicated

P10.

IF NOT BORN IN US - Length of time
in US (specify# days, months or years)

Not Indicated

P11.

Length of time in California
(specify# days, months or years)

Not Indicated

P12.

Currently Employed?

Not Indicated

P13.

Last Occupation

Not Indicated

P14.

Last grade completed in school

Not Indicated

P15.

Marital Status

P16.

Not Indicated

P17.

Family Household Income
(state as soecificallv as possible)
Welfare Eligible?

P18.

Substance Abuse?

Not Indicated

P18a.

If yes,· what substance

p.other

Married
Remarried, if so what#
D Single, never married
D Separated
Divorced
Widowed
Other
Not Indicated

p.eng
p.plang

p.pob
p.bmusa
p.tmeusa

p.tmeca

p.cremp
p.lstjob
p.educ
p.mstat

Not Indicated

p.inc
p.wfe
p.sbhx

p.sbhxt

Item#

P19.
P19a.
P19b.

P20.
P20a.
P21.
P21a.
P22.
P22a.

Perpetrator Characteristics (at time of incident)

Disability?
(physical, learning, etc.)
If yes, specify type

If yes, what assistance was
provided? (e.g., signer for hearing
impaired)
Developmental delays?

Mental Illness?

P23a.

If yes, specify

P25.
P25a.

Item#

P26.

Gender identity issue?

D Not Indicated

Perpetrator Characteristics (at time of incident)

p.dvdl

p.ment

p.emot
p.emott

D Not Indicated

p.bhv
p.bhvt

D Not Indicated

p.geni
p.genit

D Not Indicated

If yes, specify crime{s)

Currently Incarcerated?

p.disas

p.mentt

If yes, specify
Criminal History (outstanding warrants,
arrests, or convictions)?

p.dis

p.dvdlt

D Not Indicated

If yes, specify
Behavior problems?

P24a.

D Not Indicated

If yes, specify diagnosis
Emotional problems?

SPSS

p.dist

D No Assistance
D Not Indicated

If yes, specify type

P23.

P24.

D Not Indicated

p.crhx

p.crhxt

D Not Indicated

SPSS
p.carc

I

P26a.

If yes, specify location

I

p.carrJ

Item#

21.

21a.

22.
22a

Item#

General Family/Caretaker Characteristics (at time of incident)
Family Problems?

Not Indicated

If yes, describe: {issues related to
divorce, separation, child custody,
marital discord ... )

Domestic Violence?
If yes, describe {specify?) {in how
much detail?)

General Family/Caretaker Characteristics (at time of incident)

SPSS
z.fmprb
z.prbd

Not Indicated

z.dv
z.dvd

SPSS

23.
23a.

24.
24a.

Parenting Problems
If yes, describe (specify?) (in how
much detail?)

Income Problems

Number siblings (# of brothers and
sisters, including steps and halves)

25a.

Number of associated (?) siblings in
placement/system

26.

Generational involvement in CWS?
Specify which generations and how

z.par
z.pard

Not Indicated

If yes, describe (specify?) (in how
much detail?)

25.

26a.

Not Indicated

z.inc
z.incd

z.#sibs

z.#sys

Not Indicated

z.gnr
z.gnrd

Outcomes of Jurisdictional/Dispositional Hearing
(Attach copy of the case plan if available)
Y1.

Outcomes of Jurisdictional/Dispositional Hearing
Date(s} of the Jurisdictional/
Dispositional Hearing

Y1a.

Case has been assigned to

Family Maintenance
(child remains with family
Family Reunification
(child is removed from home}
Case has been denied services
(bypassed}
Other, please describe:

Y2.

Child's placement will be

with mother
with father
with mother and father
with alternate caretaker
specify
Family Foster Home (FFH}
Family Foster Agency (FFA}
group home
relative foster home (kincare}

Item#

SPSS
y.jddte

y.csasg

y.plmt

other
Y3.

Services ordered for the child (if any}

Item#

Outcomes of Jurisdictional/Dispositional Hearing

y.sordc

SPSS

Y4.

Services ordered for the mother {if any)

Y5.

Services ordered for the father {if any)

Y6.

Services ordered for the alternate
caretaker {if any)

Y7.

Case under provisions of the Indian
Child Welfare Act?

Item#

Outcomes of Jurisdictional/Dispositional Hearina

y.sordm

y.sordf

y.sorda

D Not Indicated

y.iwa

SPSS

Y8.

Changes that occurred between
detention and jurisdictional /
dispositional hearing

y.chg

Changes or Major Events in Case Characteristics at Subsequent Hearings
Item#

Changes in Case Characteristics
For each change, please note in the
appropriate sections:
I. Date of Hearing
II. Type of Hearing
Ill Person(s) Affected
IV Change or Major Event
Service Related
• Placement
• Living Situation
• Court Mandate
• Adherence to participation in
services
• New services rendered
• Participation in other service
systems
• Other?

H1a.

Date of Hearing

H1b.

Type of Hearing

H1c.

Person(s) Affected

H1d.

Change or Major Event

SPSS

Specify change/event and describe
condition
Psychosocial (improvement,
decline, unchanged?)
• Academic
• Substance abuse {relapse,
improvement?)
• Disability
• Developmental
• Mental Illness
• Emotional
• Behavioral
• Gender Identity
• Criminal
• Health Status
• Other?

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

H2a.

Date of Hearing

H2b.

Type of Hearing

H2c.

Person{s} Affected

H2d.

Change or Major Event

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

..

H3a.

Date of Hearing

H3b.

Type of Hearing

H3c.

Person{s} Affected

H3d.

Change or Major Event

H4a.

Date of Hearing

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

H4b.

Type of Hearing

H4c.

Person(s) Affected

H4d.

Change or Major Event

H5a.

Date of Hearing

H5b.

Type of Hearing

HSc.

Person(s) Affected

H5d.

Change or Major Event

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

H6a.

Date of Hearing

H6b.

Type of Hearing

H6c.

Person(s) Affected

H6d.

Change or Major Event

H7a.

Date of Hearing

H7b.

Type of Hearing

H7c.

Person(s) Affected

H7d.

Change or Major Event

HBa.

Date of Hearing

Mother
Father

D Family
Alternate Caretaker
Other

Specify change/event and describe condition

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

H8b.

Type of Hearing

H8c.

Person(s) Affected

H8d.

Change or Major Event

H9a.

Date of Hearing

H9b.

Type of Hearing

H9c.

Person(s) Affected

H9d.

Change or Major Event

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

H10a.

Date of Hearing

H10b.

Type of Hearing

H10c.

Person(s) Affected

H10d.

Change or Major Event

H11a.

Date of Hearing

H11b.

Type of Hearing

H11c.

Person(s) Affected

H11d.

Change or Major Event

H12a.

Date of Hearing

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

H12b.

Type of Hearing

H12c.

Person(s) Affected

H12d.

Change or Major Event

H13a.

Date of Hearing

H13b.

Type of Hearing

H13c.

Person(s) Affected

H13d.

Change or Major Event

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

Mother
Father

Family
Alternate Caretaker
Other

Specify change/event and describe condition

Notes:
Case Notes (other pertinent information ... ):

Coder Notes (impressions, problems, reactions ... ):

ATTACHMENT 4:
Reliability Study

Reliability Study

In the course of case-record review, ten different people each will be recording
information from forty cases. As each person reads a case and makes decisions about
characteristics of that case, it is important to know how similar that person's decisions are
to decisions that would have been made if another person had read the case. Put another
way, it is important to establish whether case reviewers are consistent in the judgments
they make.
Consistency among case record reviewers establishes confidence in the
information that is being obtained. Subsequent analyses of the data and interpretation are
strengthened by assurances that the underlying information is reliable.
Method. We assessed the consistency, or reliability, of case record review in the
following manner. We selected four cases for review and asked that all case record
reviewers, ten people, review each case. The reviewers were instructed to record
information gathered from each case on our standard case record review extraction form
(Attachment 4). This procedure resulted in forty completed protocols-resulting from ten
reviewers each reading and responding to four cases.
Interest centers on the level of agreement between case record reviewers as they
recorded individual items of information asked about on the case-record review protocol.
We therefore calculated, for items where it was appropriate to do so, the proportion of
agreement among raters. This translates into the following indicator: Considering the ten
reviewers, in what proportion of instances did they agree?
The scope of our calculations included Client and Family Characteristics,
Perpetrator Information, and Outcomes of Jurisdictional/Dispositional Hearing. It

included demographics and social history, characteristics of the precipitating event, and
psychosocial problems recorded as "yes", "no", or "not indicated".
Other items require open-ended or narrative recording on the part of case-record
reviewers. Information of this kind describes, in detail, criminal history and criminal
justice involvement, services provided, and changes and major events over the course of
child welfare participation. The format does not lend itself, at least at the present time, to
calculations of the kind described above. We are devising additional procedures,
described below, to categorize these responses.
Agreement was calculated as the number of times the most frequently recorded
response was recorded, divided by the number of case record reviewers--that is, ten. For
example, when judging the presence of a mental health problems, if eight reviewers
checked "yes", one checked "no" and one "not indicated", the assessed level of
agreement would be 8/10, or 80%.
In some instances, additional questions had to be answered before calculation
could proceed. Issues to be resolved include those where discrete alternatives have not
been provided-birth date, for example, or place of birth. The issue was what kind of
similarity would constitute agreement? Another issue was handling of blank spaces on
the case review protocol where the reviewer chose to write nothing. Finally, how many
case record reviewers must agree? What percentage of agreement was to be considered
acceptable?
In order to avoid overstating levels of agreement, we proceeded from a
conservative stance in answering these questions. We insisted on perfect agreement
where information was not recorded in previously provided, structured categories. For

example, we required that the match be perfect among recorded birth dates and places;
any deviation was considered a mismatch. We instituted on perfect agreement among
responses to structured item. Thus, we did not discard protocols where no information
was recorded but counted them and matched them with other protocols, if any, where no
information was recorded. To determine an acceptable level of interrater agreement, we
adopted a threshold of 80%; we required that eight out of ten case record reviewers
exactly agree to count the question as reaching a good level ofreliability.
Results. For the four cases considered in the reliability study, 64% of case record

review items were found to have a good level of inter-rater reliability. Thus, 80% of ten
reviewers recorded with perfect agreement on almost two-thirds of items assessed
included on the case record review protocol.
Discussion. Without further modification almost two thirds of questions on the

case record review protocol form can be considered, by our stringent standards, to
provide a reliable basis for recording. Items not meeting this standard have been
identified and can be improved by simple modification of procedure. In an initial step, we
fed back results from our reliability analysis to case-record reviewers and distributed
guidelines for improvement (see attached).
We supplemented our procedures for data collection in order to screen protocols
and resolve ambiguities. For example, blank items-which surpressed levels of assessed
reliability in several instances-will be followed up and needed information ascertained
and recorded. By this procedure, we expect to further strengthen levels of agreement
achieved by case record reviewers in the course of their work.

We are now at work to develop procedures for coding criminal history and
criminal justice involvement, services provided, and changes that occur over the course
of child welfare involvement. The result of our efforts, a standardized grid for
categorizing some events that take place and indicators for summarizing others, will
permit us to understand our cases at even a higher level of detail.
In addition, we will assess the reliability of case record review during the course
of the study. We will chose a sample of forty cases such that reviewers are unaware of
which cases have been selected. We will then perform a second case record review on
those cases, without consulting the protocol from the original review. Using methods
described above, we will then assess levels of agreement between the first and second
reviews. We will thereby know levels of reliability of recorded information at the heart of
our study.

ATTACHMENT 5:
Variable Checklist

Santa Clara County/ SJSU College of Social Work
Variable Checklist (Working List)
CWS-CMS & Case Record Data Extraction Form

Family Characteristics (at time of incident)
Child
Item#
C1.

Family Characteristics (at time of incident) - Child
Sex

C2.

Ethnicity

CWS-CMS

X

Case Review

X

X

Primary Ethnicity

X

Secondary Ethnicity

X

C2.

Speaks English?

X

C2.

Translator needed? Indicate language

X

C3.

C4.

cs.
C6.

Primary Language

X

Secondary Language

X

Type(s) of abuse/neglect,
Sustained bv court in minute order
Reason for opening case

X

Number of prior referrals

X

X

Initial Reporter of Incident (police
report: reporting party, redacted)

X

Date of incident

X

X

Date of case closing

X

C7.

Date of birth

X

ca.

Born in US?

C9.

Place of Birth {city, state, country)

C10.

Length of time in US
(soecifv # days, months or years}
Length of time in California
(soecifv # davs, months or vears)
Last grade completed in school

C12.
C13.

C13a.

Academic problems {including truancy
and dropped out)
If yes, specify type

c.sex
c.eth

c.enQ
c.plang

c.rsn4rf

c.or#ref
c.pr#cws

c.reptr

X

Date of case opening

C11.

X

Previous # times in CWS
{recipient of CWS services)

Reporter of Incident
C6a.

X

SPSS

X
X

X

X
X
X

X

X
X
X

c.datei

c.bdav
c.brnusa
c.pob
c.tmeusa
c.tmeca
c.educ
c.acpb

c.acobt

C14.
C14a.
C15.

Substance Abuse?
If yes, what substance

X
X

C15a.

Disability?
(phvsical, learning, etc.)
If yes, specify type

X

C16.

Developmental delays?

X

C16a.

If yes, specify type

C17.
C17a.
C18.
C18a.

Mental Illness?
If yes, specify diagnosis
Emotional problems?
If yes, specify

X

X
X
X
X
X

C19.

Behavior problems?

X

C19a.

If yes, specify

X

C20.
C20a.
C21
C21a.

Gender identity issue?
If yes, specify
Criminal History (arrest or conviction)?
If yes, specify crime(s)

X
X
X
X

C22.

Current health status

X

C23.

Participation in other service s~stems?

X

If yes, specify ALL service systems

X

C23a.
C24.

C24a.

C25.

Participation in other services?
If yes, specify ALL services, (e.g.
child advocacy, mentorship,
counseling ... )
Child's current living situation

X
X

X

c.sbhx
c.sbhxt
c.dis
c.dist
c.dvdl
c.dvdlt
c.ment
c.mentt
c.emot
c.emott
c.bhv
c.bhvt
c.aeni
c.aenit
c.crhx
c.crhxt
c.hths
c.ptsv
c.ptsvt
C.POSV

c.posvt

c.crls

Family Characteristics (at time of incident)
Mother
Item#

M1.

Familv Characteristics (at time of incident) - Mother

Ethnicity

CWS-CMS

Case Review

X

Primary Ethnicity

X

Secondary Ethnicity

X

M2.

Date of Birth

X

M3.

Speaks English?

X

M4.

Translator needed? Indicate language

X

Primary Language

X

Secondary Language

X

X

MS.

Place of Birth (city, state, country)

X

M6.

Born in US?

X

M7.

X

M9.

Length of time in US
(specify# davs, months or vears)
Length of time in California
(specify# davs, months or years)
Currently Employed?

M10.

Last Occupation

X

M11.

Last grade completed in school

X

M12.

Marital Status

X

M13.

X

M14.

Family Household Income
(state as specificallv as possible)
Welfare Eligible?

X

M15.

Substance Abuse?

X

MB.

M15a.

If yes, what substance

X
X

X
X

M16a.

Disability?
(physical, learning, etc.)
If yes, specify type

M17.

Developmental delays?

X

If yes, specify type

X

M16.

M17a.
M18.
M18a.
M19.

Mental Illness?
If yes, specify diagnosis
Emotional problems?

X

X
X
X

If yes, specify

X

M20.

Behavior problems?

X

M21a.

If yes, specify

X

M19a.

SPSS

m.eth

m.bdav
m.enQ
m.plang

m.oob
m.bmusa
m.tmeusa
m.tmeca
m.cremp
m.lstiob
m.educ
m.mstat
m.inc
m.wfe
m.sbhx
m.sbhxt
m.dis
m.dist
m.dvdl
m.dvdlt
m.ment
m.mentt
m.emot
m.emott
m.bhv
m.bhvt

M22.
M22a.
M23.
M23a.
M24.
M24a.

Gender identity issue?
If yes, specify
Criminal History (arrest or conviction)?
If yes, specify crime(s)
Currently Incarcerated?
If yes, specify location

X
X
X
X
X
X

M25.

Current health status

X

M26.

Participation in other service systems?

X

If yes, specify ALL service systems

X

M26a.
M27.

Participation in other services?

X

M27a.

If yes, specify ALL services, (e.g.
Counseling, anger management,
ParentinQ class ... )
Mother's current living situation
(check all that apply)
Is mother the child's primary caretaker?

X

Does mother currently (do we mean at time of referral?)
have legal custody of child?

X

M28.
M29.
M30.

X
X

m.geni
m.genit
m.crhx
m.crhxt
m.carc
m.carcl
m.hths
m.ptsv
m.ptsvt
m.posv
m.posvt
m.crls
m.orctk
m.cust

Family Characteristics (at time of incident)
Father
Item#
F1.

Family Characteristics lat time of incident) - Father
Ethnicity

CWS-CMS

Primary Ethnicity

X

Secondary Ethnicity

X

Case Review

X

F2.

Date of Birth

X

F3.

Speaks English?

X

F4.

Translator needed? Indicate language

X

Primary Language

X

Secondary Language

X

F5.

Place of Birth (city, state, country)

X

F6.

Born in US?

X

F7.

X

F9.

Length of time in US
(specify# days, months or years)
Length of time in California
(specify# days, months or vears)
Currently Employed?

X

F10.

Last Occupation

X

F11.

Last grade completed in school

X

F12.

Marital Status

X

F13.

X

F14.

Family Household Income
(state as specificallv as oossible)
Welfare Eligible?

F15.

Substance Abuse?

X

F8.

F15a.

If yes, what substance

X

X

X
X

F16a.

Disability?
(physical, learning, etc.)
If yes, specify type

F17.

Developmental delays?

X

If yes, specify type

X

F16.

F17a.
F18.
F18a.
F19.
F19a.
F20.
F21a.

Mental Illness?
If yes, specify diagnosis
Emotional problems?

X

X
X
X

If yes, specify

X

Behavior problems?

X

If yes, specify

X

SPSS

f.eth

f.bday
f.eng
f.plang

f.pob
f.bmusa
f.tmeusa
f.tmeca
f.cremp
f.lstjob
f.educ
f.mstat
f.inc
f.wfe
f.sbhx
f.sbhxt
f.dis
f.dist
f.dvdl
f.dvdlt
f.ment
f.mentt
f.emot
f.emott
f.bhv
f.bhvt

F22.
F22a.
F23.
F23a.
F24.
F24a.

Gender identity issue?
If yes, specify
Criminal History (arrest or conviction)?
If yes, specify crime(s)
Currently Incarcerated?
If yes, specify location

X
X
X
X
X
X

F25.

Current health status

X

F26.

Participation in other service systems?

X

If yes, specify ALL service systems

X

F26a.
F27.

Participation in other services?

X

F27a.

X

F28.

If yes, specify ALL services, (e.g.
Counseling, anger management,
Parentirn:i class ... )
Father's current living situation

X

F29.

Is father the child's primary caretaker?

X

F30.

Does father currently (do we mean at time of referral?)
have leaal custody of child?

X

f.aeni
f.oenit
f.crhx
f.crhxt
f.carc
f.carcl
f.hths
f.otsv
f.ptsvt
f.oosv
f.posvt
f.crls
f.orctk
f.cust

Family Characteristics (at time of incident)
Alternate Caretaker (Assess if neither mother or father is the primary caretaker)
Item#
A1.
A2.

Family Characteristics (at time of incident) Alternate Caretaker
Relationship to child

CWS-CMS

Case Review

X
X

A3.

Length of time as caretaker
{soecifv # days, months or years)
Sex

X

A4.

Ethnicity

X

Primary Ethnicity

?

Secondary Ethnicity

?

A5.

Date of Birth

?

AS.

Speaks English?

X

A7.

Translator needed? Indicate language

X

Primary Language

?

Secondary Language

?

X

A8.

Place of Birth (city, state, country)

X

A9.

Born in US?

X

A10.

X

A12.

Length of time in US
(soecifv # davs, months or years)
Length of time in California
(specifv # days, months or years)
Currently Employed?

X

A13.

Last Occupation

X

A14.

Last grade completed in school

X

A15.

Marital Status

X

A16.

X

A17.

Family Household Income
(state as specifically as possible)
Welfare Eligible?

X

A18.

Substance Abuse?

X

A11.

A18a.

If yes, what substance

X

X
X

A19a.

Disability?
(physical, learning, etc.)
If yes, specify type

A20.

Developmental delays?

X

If yes, specify type

X

A19.

A20a.
A21.
A21a.

Mental Illness?
If yes, specify diagnosis

X

X
X

SPSS

a.rel2ch
a.tmasct
a.sex
a.eth

a.bdav
a.enQ
a.plang

a.oob
a.brnusa
a.tmeusa
a.tmeca
a.cremo
a.lstiob
a.educ
a.mstat
a.inc
a.wfe
a.sbhx
a.sbhxt
a.dis
a.dist
a.dvdl
a.dvdlt
a.ment
a.mentt

A22.
A22a.

Emotional problems?

X

If yes, specify

X

A23.

Behavior problems?

X

A23a.

If yes, specify

X

A24.
A24a.
A25.
A25a.
A26.
A26a.

Gender identity issue?
If yes, specify
Criminal History (arrest or conviction)?
If yes, specify crime(s)
Currently Incarcerated?
If yes, specify location

X
X
X
X
X
X

A27.

Current health status

X

A28.

Participation in other service s~stems?

X

If yes, specify ALL service systems

X

A28a.
A29.

Participation in other services?

X

A29a.

X

A30.

If yes, specify ALL services, (e.g.
Counseling, anger management,
Parenting class ... )
Alternate caretaker's current living situation

A31.

Is alternate caretaker the child's primary caretaker?

X

X

a.emot
a.emott
a.bhv
a.bhvt
a.aeni
a.aenit
a.crhx
a.crhxt
a.care
a.carcl
a.hths
a.ptsv
a.ptsvt
a.posv
a.posvt
a.eris
a.prctk

Perpetrator Information (at time of incident)
Item#
P1.
P2.

Perpetrator Characteristics (at time of incident)
Perpetrator (relationship to child)

CWS-CMS

Case Review

X

Is the perpetrator someone other than the mother,
father, or alternate caretaker?

X

P3.

Sex

X

P4.

Ethnicity

X

Primary Ethnicity

SPSS
p.rel2ch
p.other
p.sex
o.eth

Secondary Ethnicity
PS.

Date of Birth

X

P6.

Speaks English?

X

P7.

Translator needed? Indicate language

X

Primary Language

p.bday
p.eng
p.plang

Secondary Language
PB.

Place of Birth (city, state, country)

X

P9.

Born in US?

X

P10.

X
X

P12.

Length of time in US
(specify# days, months or years)
Length of time in California
(specify# days, months or years)
Currently Employed?

P13.

Last Occupation

X

P14.

Last grade completed in school

X

P15.

Marital Status

X

P16.

X

P17.

Family Household Income
(state as specifically as oossible)
Welfare Eligible?

P18.

Substance Abuse?

X

P11.

P18a.

If yes. what substance

X

X

X
X

P19a.

Disability?
{physical, learning, etc.)
If yes, specify type

X

P20.

Developmental delays?

X

If yes, specify type

X

P19.

P20a.
P21.
P21a.
P22.

Mental Illness?
If yes, specify diagnosis
Emotional problems?

X
X
X

p.pob
p.brnusa
p.tmeusa
p.tmeca
p.cremp
o.lstiob
o.educ
o.mstat
p.inc
p.wfe
p.sbhx
p.sbhxt
p.dis
p.dist
p.dvdl
o.dvdlt
p.ment
p.mentt
p.emot

P22a.

If yes, specify

X

P23.

Behavior problems?

X

P23a.

If yes, specify

X

P24.

P24a.
P25.
P25a.
P26.
P26a.

Gender identity issue?
If yes, specify
Criminal History (arrest or conviction)?
If yes, specify crime(s)
Currently Incarcerated?
If yes, specify location

X
X
X
X
X
X

p.emott
o.bhv
P.bhvt
p.geni
p.genit
p.crhx
o.crhxt
o.carc
o.carcl

Item#
21.
21a.

22.
22a
23.
23a.
24.
24a.

General Family/Caretaker Characteristics (at time of
incident)
Family Structure Problems?
If yes, describe: (issues related to
divorce, separation, child custody,
marital discord ... )
Domestic Violence?
If yes, describe (specify?) (in how much detail?)
Parenting Problems
If yes, describe (specify?) (in how much detail?)
Income Problems
If yes, describe (specify?) (in how much detail?)

CWS-CMS

Case Review

X
X
X
X
X
X
X
X

25.

Number siblings

X

25a.

Number of associated (?) siblings in placement/system

X

26.

Generational involvement in CWS?

X

26a.

Specify which generations and how

X

SPSS

z.fmprb
z.prbd

z.dv
z.dvd
z.oar
z.oard
z.inc
z.incd
z.#sibs
z.#svs
z.anr
z.anrd

Outcomes of Jurisdictional/Dispositional Hearing
(Attach copy of the case plan if available)
Item#
Y1.

Outcomes of Jurisdictional/Dispositional Hearing

Y2.

Child's placement will be (with whom or service type)

X

Y3.

Services ordered for the child (if any)

X

Y4.

Services ordered for the mother (if any)

X

Y5.

Services ordered for the father (if any)

X

Y6.

Services ordered for the alternate caretaker (if any)

X

Y7.

Case under provisions of the Indian Child Welfare Act?

X

Case has been assigned to (FM, FR, Bypass, Other)

CWS-CMS

?

Case Review

X

SPSS

y.csasg
v.olmt
y.sordc
y.sordm
y.sordf
y.sorda
y.iwa

Changes or Major Events in Case Characteristics at Subsequent Hearings
Child
Item#

Changes/Major Events in Case Characteristics Child
For each change, please note in the appropriate sections:

CWS-CMS

Case Review

SPSS

X

I. Date of Hearing
II. Type of Hearing
Ill. Change or Major Event

Changes or Major Events in Case Characteristics at Subsequent Hearings
Mother
Item#

Changes/Major Events in Case Characteristics Mother
For each change, please note in the appropriate sections:

CWS-CMS

Case Review

SPSS

X

I. Date of Hearing
II. Type of Hearing
Ill. Change or Major Event

Changes or Major Events in Case Characteristics at Subsequent Hearings
Father
Item#

Changes/Major Events in Case CharacteristicsFather
For each change, please note in the appropriate sections:

CWS-CMS

Case Review

SPSS

X

I. Date of Hearing
II. Type of Hearing
Ill. Change or Major Event

Changes in or Major Events Case Characteristics at Subsequent Hearings
Alternate Caretaker
Item#

Changes/Major Events in Case Characteristics Father
For each change, please note in the appropriate sections:

CWS-CMS

Case Review

SPSS

X

I. Date of Hearing
II. Type of Hearing
Ill. Change or Major Event

Changes in or Major Events Case Characteristics at Subsequent Hearings
General Family/Caretaker Characteristics
Item#

Changes or Major Events in General Family/Caretaker CWS-CMS
Characteristics
For each change. please note in the appropriate sections:
I. Date of Hearing
II. Type of Hearing
Ill. Change or Major Event

Case Review

X

SPSS

Emergency Response Screening
Reason for case closure
Voluntary placement (type)

CWS-CMS

X

Case Review

SPSS

Case Review

I SPSS

X

Reason for case dismissal (2°0 phase only examining
cases substantiated by the court)
Reason case opened for services (same as
abuse/neglect substantiated variable)
Supervisor Characteristics
Ethnicity
Training

CWS-CMS

l

From personnel records?

# years in position

Caseworker Characteristics
Ethnicity
Training

CWS-CMS

l

Case Review

I SPSS

From personnel records?

# years in position

Language ability
X

Caseworker characteristics (changes over case)

Foster care Characteristics (some available in CMS)
Type of placement (covered above)

CWS-CMS

X

Length of time in placement

X

Provider Characteristics

X

Case Review

X

SPSS

X

Type of funding
Hearings and Timelines (6 month intervals)
Hearings and case changes/major events are covered
above
Other case characteristics

CWS-CMS

Case Review

X

SPSS

Actual time to closure (calculated from CMS)

CWS-CMS
X

Case Review

# of case openings

X

# of placement changes

X

X

Placement sequences by date

X

X

# of caseworker changes

X

X

SPSS

Notes:
Researchers Case Notes (other pertinent information ... ):
Coder Notes (impressions, problems, reactions ... ):

CWS-CMS

Case Review

X

X

SPSS

ATTACHMENT 6:
Sampling Summary:
Case Availability Status Analysis

Sampling Summary: Case Availability Status Analysis
The following steps were used to create the minimum sample of 400 cases for the
purposes of in-depth case record review. First, all the CWS cases closed between
January 1, 2000 and June 30, 2001 were included in the initial sampling frame. Second,
given that siblings within a family of a CWS involved child could be in this list, a single
member within each distinct household (determined by the same identified mother) was
randomly sampled so that only one representative from each household was left in the
sampling frame. The resulting general sample size was 1753 cases. Third, random
stratified sampling was initially used to select a set of 403 cases from this set of CWS
cases. The criteria for stratification were ethnicity (recoded into 5 subgroups: Black,
Latino, White, Asian, and Other), service type identified at the close of the case (either
ER, PP, or FR versus FM), and age group (0 to 6 years, 7 to 13 years, and older than 13
years old). These strata were created in order to obtain as representative a sample as
possible reflecting the cases managed by Santa Clara County's CWS. Fourth, since we
were unable to meet the target sample size through true random stratification due to a
large number of missing case files across strata and the number ofVFM cases excluded,
we had to use purposive sampling in addition to random stratification to meet our
minimum sample size. Thus, approximately two-thirds of the way through the sampling
and data extraction phases, cases within strata with few children (i.e., Asians and Blacks)
had to be selected purposefully in order to reach the minimum 400 caseload sample and
improve representation.
The availability status of the cases originally within the sampling frame (missing
case file, voluntary family maintenance, or includable and coded) was related to the
child's ethnicity. A total of 1298 cases had to be requested from Santa Clara County's
Social Services Agency in order to obtain 403 codable cases for the case record review.
Of these 1298 cases, 529 (40.8%) were missing and/or unavailable for coding, 366
(28.2%) were identified as voluntary family maintenance, and 403(31.0%) available and
coded for the case record review component. A comparatively higher proportion of
Latino cases requested for review were missing (47.5%). A comparatively higher
proportion of Asian/PI cases requested for review were assigned to voluntary family
maintenance (52.2%). Please see Table 1.

Table 1: Case Availability Status By Ethnicitya
Total

Frequencies and % within Ethnic Subgroup
Black
White
Asian/PI
Latino

Other

Missing/Unavailable

529
(40.8% of
requested)

77
(40.7%)

161
(39.1%)

242
(47.5%)

40
(25.2%)

9
(32.1%)

Voluntary Family
Maintenance

366
(28.2% of
requested)

53
(28.0%)

103
(25.0%)

120
(23.5%)

83
(52.2%)

7
(25.0%)

Available and
Codedb

403
(31.0% of
requested)

59
(14.6%)

148
(35.9%)

148
(29.0%)

36
(22.6%)

12
(42.9%)

Based on 1298 cases requested from SCC-SSA and CWS-CMS ethnicity codes
b These proportions by ethnicity may differ from that reported in the detailed summaries of the
403 case record reviews due to re-clarification of ethnic identity found during the case review
a

ATTACHMENT 7:
Table 1:

Demographic and System Characteristics of Children,
Santa Clara County Open Cases, December 2000

Table 2:

Demographic Characteristics for Children in Out of Home
Placement (OHP), Santa Clara County Open Cases,
December 2000

Table 3:

Case Characteristics for Children in Out of Home
Placement (OHP), Santa Clara County Open Cases,
December 2000

Table 4:

Demographic and System Characteristics by Race/Ethnicity
in Out of Home Placement (OHP), Santa Clara County
Open Cases, December 2000

Table 1:

Demographic and System Characteristics of Children, Santa Clara County Open Cases, December 2000

Characteristic

N = 4399

Valid%
(% based on available data)

Race/Ethnicity
African American
White
Hispanic / Latino
Asian American I Pacific Islander
Other
Missing / Not Recorded

531
1178
2288
318
70
14

12.1%
26.9%
52.2%
7.2%
1.6%
not included

Gender
Females
Males
Missing/Not Recorded

2237
2155
7

50.9
49.1
not included

Primary Language
English
Asian/Pacific Islander Language
Spanish
Other
Missing/Not Recorded

3786
120
465
15
13

86.3%
2.7%
10.6%
0.3%
not included

Born in California
In State
Out of State
Missing/Not Recorded

1296
72
3031

94.7%
5.3%
not included

X = 9.22
sd = 5.16

Not Applicable

250
1328
868
1953

5.7%
30.2%
19.7%
44.4%

Age (in years)
Service Component
[ER] Emergency Response
[FM] Family Maintenance
[FR] Family Reunification
[PP] Permanent Placement

Table 2:

Demographic Characteristics for Children in Out of Home Placement [OHP],
Santa Clara County Open Cases, o·ecember 2000
Characteristic

N = 2721

Valid%

(% based on available data)

Race/Ethnicity
African American
White
Hispanic / Latino
Asian American / Pacific Islander
Other
Missing / Not Recorded

399
739
1381
159
41
2

14.7%
27.2%
50.8%
5.8%
1.5%
0.1%

Gender
Females
Males

1356
1365

49.8%
50.2%

Primary Language
English
Asian / Pacific Islander Language
Spanish
Other

2443
54
213
11

89.8%
2.0%
7.8%
0.4%

Born in California
In State
Out of State
Missing/Not Recorded

1040
59
1622

94.6%
5.4%
not included

=9.27
sd =5.18

Not Applicable

146
112
700
1763

5.4%
4.1%
25.7%
64.8%

Age (in years)

Service Component
[ER] Emergency Response
[FM] Family Maintenance
[FR] Family Reunification
[PP] Permanent Placement

-

X

Table 3:

Case Characteristics for Children in Out of Home Placement [OHP],
Santa Clara County Open Cases, December 2000
Characteristic

N = 2721

Reason for Removal

Caretaker Absence/ Incapacity
Emotional Abuse
General Neglect
Physical Abuse
Severe Neglect
Sexual Abuse
Voluntary Placement
Other

Valid%
(% based on available data)

1025
91
472
444
501
143
17
28

37.7%
3.3%
17.3%
16.3%
18.4%
5.3%
0.6%
1.0%

109
456
189
290
1050
627

4.0%
16.8%
6.9%
10.7%
38.6%
23.0%

=3.83
sd =3.25
X =39.97
sd =39.57

Not Applicable

Placement Home Facility Type

Small Family/Court Specified/Tribe Specified/
Guardian Home
Foster Family Home
Group Home
County Shelter
Relative Home
Foster Family Agency
Number of Placements
Number of Months in Placement

-

X

Not Applicable

Table 4a: Demographic and System Characteristics by Race/Ethnicity of Children in Out of Home Placement [OHP],
Santa Clara County Open Cases, December 2000
White
n=739

Hispanic/
Latino
n = 1381

Asian Am./
Pacific Is.
n = 159

n =41

9.86
(4.84)

9.73
(5.16)

8.93
(5.24)

8.68
(5.17)

9.07
(5.17)

Emergency
Response

13
(3.3%)

52
(7.0%)

67
(4.9%)

9
(5.7%)

5
(12.2%)

Family
Maintenance

7
(1.8%)

24
(3.2%)

69
(5.0%)

11
(6.9%)

1
(2.4%)

Family
Reunification

68
(17.0%)

196
(26.5%)

383
(27.7%)

38
(23.9%)

13
(31.7%)

Permanent
Placement

311
(77.9%)

467
(63.2%)

862
(62.4%)

101
(63.5%)

22
(53.7%)

Number of
Placements,

3.93
(3.66)

4.11
(3.55)

3.74
(3.02)

2.92
(1.84)

4.76
(4.49)

Months in
Placement,

55.32
(45.26)

38.15
(35.59)

37.87
(39.98)

29.63
(29.03)

35.55
(34.13)

Age in years,
mean (sd)

African
American
n = 399

Other

Service
Component

mean (sd)

mean (sd)

Table 4b: Demographic and System Characteristics by Race/Ethnicity of Children in Out of Home Placement [OHP],
Santa Clara County Open Cases, December 2000 (continued)

n=739

Hispanic/
Latino
n = 1381

Asian Am./
Pacific Is.
n = 159

n =41

19
(4.8%)

31
(4.2%)

47
(3.4%)

7
(4.4%)

40
(9.8%)

66
(16.5%)

134
(18.1%)

213
(15.4%)

(21.4%)

34

9
(22.0%)

Group Home

25
(6.3%)

92
(12.4%)

65
(4.7%)

5
(3.1%)

2
(4.9%)

County Shelter

27
(6.8%)

84
(11.4%}

159
(11.5%}

15
(9.4%}

5
(12.2%)

Relative Home

172
(43.1%)

222
(30.0%)

583
{42.2%)

60
(37.7%)

13
(31.7%)

Foster Family Agency

90
(22.6%)

176
(23.8%)

314
(22.7%)

38
(23.9%)

8
(19.5%)

African
American
n=399

Small Family/
Court Spec.I
Tribe Spec.I
Guardian Home
Foster Family Home

White

Other

Placement
Facility Type

Table 4c: Demographic and System Characteristics by Race/Ethnicity of Children in Out of Home Placement [OHP],
Santa Clara County Open Cases, December 2000 (continued)
African
American
n=399

Caretaker
Absence/
Incapacity

White
n=739

Hispanic/
Latino
n 1381

Asian Am./
Pacific Is.
n 159

n=41

177
(44.4%)

276
(37.3%)

511
(37.0%)

50
(31.4%)

11
(26.8%)

Emotional
Abuse

12
(3.0%)

27
(3.6%)

42
(3.0%)

9
(5.7%)

0
(0.0%)

General
Neglect

63
(15.8%)

143
(19.3%)

235
(17.0%)

24
(15.1%)

7
(17.1%)

Physical
Abuse

56
(14.0%)

124
(16.8%)

218
(15.8%)

37
(23.3%)

9
(22.0%)

Severe
Neglect

68
(17.0%)

110
(14.9%)

286
(20.7%)

27
(17.0%)

5
(12.2%)

Sexual Abuse

19
(4.8%)

40
(5.4%)

68
(4.9%)

11
(6.9%)

0
(0.0%)

Voluntary
Placement

1
(0.3%)

8
(1.1%)

8
(0.3%)

1
(0.6%)

0
(0.0%)

Other

3
(0.8%)

12
(1.6%)

13
(0.9%)

0
(0.0%)

0
(0.0%)

=

=

Other

Reason for
Removal