Foster Care: Kinship Care Quality and Permanency Issues (Letter Report,
05/06/99, GAO/HEHS-99-32).

Pursuant to a congressional request, GAO reviewed how well kinship care
is serving foster children, focusing on the: (1) quality of care that
children in kinship care receive compared with that received by other
foster children, as measured by a caseworker's assessment of a
caregiver's parenting skills, the extent to which a foster child is able
to maintain contact with familiar people and surroundings, and a
caregiver's willingness to enforce court-ordered restrictions on
parental visits; (2) frequency with which state child welfare agencies
pursue various permanent living arrangements and the time children in
kinship care have spent in the system compared with other foster
children; and (3) recent state initiatives intended to help ensure that
children in kinship care receive good quality foster care and are placed
in permanent homes in a timely manner.

GAO noted that: (1) GAO's survey of open foster care cases in California
and Illinois showed that in most respects the quality of both kinship
and other foster care was good and that the experiences of children in
kinship care and children in other foster care settings were comparable;
(2) GAO found that caregivers both in kinship care and in other foster
care settings demonstrated good parenting skills overall; (3) GAO also
confirmed the belief that there is more continuity in the lives of
children in kinship care before and after they enter foster care than
there is in other foster children's lives; (4) in cases in which the
courts have restricted parental visits with foster children to ensure
the children's safety, the proportion of cases in which the caseworker
believed that the caregiver was likely to enforce the restrictions was
somewhat smaller among kinship care cases than among other foster care
cases; (5) some of the standards that California and Illinois use to
ensure good quality foster care and the level of support each state
provides to foster caregivers are lower for kinship care than other
types of foster care; (6) previous research on children who have left
foster care has shown that children who had been in kinship care were
less likely to be adopted and stayed longer in foster care than other
foster children; (7) between California and Illinois, GAO's survey
showed no consistent findings regarding the relationship between kinship
care and permanency goals or the time foster children had spent in the
system; (8) in Illinois, kinship care cases were more likely to have a
permanency goal of adoption or guardianship than other foster care
cases; (9) Illinois has found that kinship caregivers are willing to
adopt, and Illinois is actively pursuing adoption in kinship care cases;
(10) in California, kinship care cases were less likely than other
foster care cases to have adoption or guardianship as a goal; (11)
according to California officials, this may be because, at the time of
GAO's survey, the state had only recently begun to offer adoption and
guardianship options specifically designed for a foster child's
relatives; (12) in California, there was no significant difference
between the average length of time that children in kinship care and
children in other settings had spent in the system; (13) in Illinois,
children in kinship care had spent significantly less time in the system
than other foster children; and (14) both California and Illinois are
now taking steps to better ensure the good quality of kinship care and
to encourage kinship caregivers to provide permanent homes for foster
children who cannot return to their parents.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  HEHS-99-32
     TITLE:  Foster Care: Kinship Care Quality and Permanency Issues
      DATE:  05/06/99
   SUBJECT:  Foster children
	     Comparative analysis
	     Families
	     Child custody
	     Child adoption
	     State programs
	     Guardians
IDENTIFIER:  California
	     Illinois
	     HHS Temporary Assistance for Needy Families Program
	     Title IV-E Foster Care Program
	     Aid to Families with Dependent Children Program
	     AFDC

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  This text was extracted from a PDF file.        **
** Delineations within the text indicating chapter titles,      **
** headings, and bullets have not been preserved, and in some   **
** cases heading text has been incorrectly merged into          **
** body text in the adjacent column.  Graphic images have       **
** not been reproduced, but figure captions are included.       **
** Tables are included, but column deliniations have not been   **
** preserved.                                                   **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************
FOSTER CARE: Kinship Care Quality and Permanency Issues GAO/HEHS-
99-32 United States General Accounting Office

GAO Report to the Chairman, Subcommittee on Human Resources,
Committee on

Ways and Means, House of Representatives

May 1999 FOSTER CARE Kinship Care Quality and Permanency Issues

GAO/HEHS-99-32

  GAO/HEHS-99-32

GAO United States General Accounting Office

Washington, D. C. 20548 Health, Education, and Human Services
Division

B-279199 May 6, 1999 The Honorable Nancy L. Johnson Chairman,
Subcommittee on Human Resources Committee on Ways and Means House
of Representatives

Dear Ms. Johnson: The Personal Responsibility and Work Opportunity
Reconciliation Act of 1996 (P. L. 104- 193) required the states to
consider giving priority to relatives when deciding where to place
the quarter of a million children who enter foster care each year
while they are in the child welfare system. In 1995, an estimated
25 percent of all foster children were living with their
relatives. Many child welfare experts believe that placing foster
children with relatives, a practice commonly known as kinship
care, can be beneficial to many of the children. Research has
shown, however, that foster children in kinship care may not
always receive good quality care, remain in the system longer than
other foster children, and are less likely to find a permanent
home outside the foster care system when they cannot return to
their parents. These findings are especially significant in light
of the Adoption and Safe Families Act of 1997 (P. L. 105- 89),
which includes provisions to ensure foster children's safety and
to speed up the process for finding permanent homes for them when
they cannot return to their parents.

In response to the request of the previous chairman of the
subcommittee for information on how well kinship care is serving
foster children, this report describes (1) the quality of care
that children in kinship care receive compared with that received
by other foster children, as measured by a caseworker's assessment
of a caregiver's parenting skills, the extent to which a foster
child is able to maintain contact with familiar people and
surroundings, and a caregiver's willingness to enforce court-
ordered restrictions on parental visits; (2) the frequency with
which state child welfare agencies pursue various permanent living
arrangements (that is, permanency planning goals) and the time
children in kinship care have spent in the system compared with
other foster children; and (3) recent state initiatives intended
to help ensure that children in kinship care receive good quality
foster care and are placed in permanent homes in a timely manner.

GAO/HEHS-99-32 Kinship Foster Care Page 1

B-279199

In conducting this work, we reviewed recent research, federal
statutes and regulations, and California and Illinois legislation
and initiatives regarding kinship care. In addition, we surveyed
samples of foster care cases in California and Illinois that were
in their foster care systems on September 15, 1997, and had been
there since at least March 1, 1997. We selected these two states
because they have large kinship care populations, have different
child welfare administration structures, and are located in
different geographic areas. We asked the caseworker responsible
for each case to respond to a questionnaire regarding several
dimensions of the quality of foster care in that case and the
permanency goals pursued as of September 15, 1997. Because this
survey is limited to the foster care population in two states, the
results cannot be generalized to the foster care population either
nationwide or in any other individual state. However, results can
be generalized to these two states, which account for about one-
quarter of the nation's foster care population and almost half of
the kinship care population nationwide. We conducted our fieldwork
between April 1997 and December 1998 in accordance with generally
accepted government auditing standards. A more detailed
description of our scope and methodology appears in appendix I.

Results in Brief Our survey of open foster care cases in
California and Illinois showed that in most respects the quality
of both kinship and other foster care was good

and that the experiences of children in kinship care and children
in other foster care settings were comparable. We found that
caregivers both in kinship care and in other foster care settings
demonstrated good parenting skills overall. We also confirmed the
generally held belief that there is more continuity in the lives
of children in kinship care before and after they enter foster
care than there is in other foster children's lives. However, in
cases in which the courts have restricted parental visits with
foster children to help ensure the children's safety, the
proportion of cases in which the caseworker believed that the
caregiver was likely to enforce the restrictions was somewhat
smaller among kinship care cases than among other foster care
cases. Moreover, some of the standards that California and
Illinois use to ensure good quality foster care and the level of
support each state provides to foster caregivers are lower for
kinship care than other types of foster care.

Previous research on children who have left foster care has shown
that children who had been in kinship care were less likely to be
adopted and stayed longer in foster care than other foster
children. Between California and Illinois, our survey showed no
consistent findings regarding the

GAO/HEHS-99-32 Kinship Foster Care Page 2

B-279199

relationship between kinship care and permanency goals or the time
foster children had spent in the system. In Illinois, kinship care
cases were more likely to have a permanency goal of adoption or
guardianship than other foster care cases. Illinois has found
that, contrary to popular belief, kinship caregivers are willing
to adopt, and Illinois is actively pursuing adoption in kinship
care cases. In California, in contrast, kinship care cases were
less likely than other foster care cases to have adoption or
guardianship as a goal. According to California officials, this
may be because, at the time of our survey, the state had only
recently begun to offer adoption and guardianship options
specifically designed for a foster child's relatives. We
calculated the length of time foster children in our survey had
been in the system as of September 15, 1997. In California there
was no significant difference between the average length of time
that children in kinship care and children in other settings had
spent in the system. In Illinois, children in kinship care had
spent significantly less time in the system than other foster
children. Nevertheless, more than 80 percent of the children in
kinship care in each state had been in care longer than the
maximum period of time generally allowed by the Adoption and Safe
Families Act of 1997 (which was enacted after the period covered
by our survey) before a state would be required to initiate
procedures to terminate parental rights.

Both California and Illinois are now taking steps to better ensure
the good quality of kinship care and to encourage kinship
caregivers to provide permanent homes for foster children who
cannot return to their parents. Both states are attempting to
enlarge the pool of potential kinship caregivers, applying more
stringent standards and approval criteria for kinship caregivers,
and providing them with support services such as counseling and
respite care. In addition, these states are using kinship adoption
and guardianship with continued maintenance payments to secure
permanent homes for foster children outside the foster care
system.

Background The Department of Health and Human Services (HHS) is
responsible for the administration and oversight of federal
funding to states for services to

foster children under title IV- E of the Social Security Act. The
states are responsible for administering foster care programs,
which are supported in part with federal funds. These funds
reimburse the states for a portion of the cost of maintaining
foster children whose parents meet federal eligibility criteria
for the funds. The criteria are based in part on the income level
of the parents. Federal expenditures for the administration

GAO/HEHS-99-32 Kinship Foster Care Page 3

B-279199

and maintenance of foster care cases eligible for title IV- E were
$3.2 billion in 1997. When foster children are not eligible for
title IV- E funding, they may be eligible for child- only benefits
under the Temporary Assistance for Needy Families (TANF) program,
which are partially funded by the federal government. Otherwise,
states and counties must bear the full cost of caring for foster
children. 1

Within the foster care system, children can be placed in any of a
number of temporary settings, including kinship care, family
foster care, private for- profit or nonprofit child care
facilities, or public child care institutional care. In the
kinship care setting, foster children are placed with their
relatives. While the definition of relatives varies somewhat by
state, relatives are typically adults who are related to a foster
child by blood or marriage. They may also be family friends,
neighbors, or other adults with whom the child is familiar. In
this report, kinship care refers to the formal placement of
children in the foster care system with their relatives. It does
not include informal arrangements for relatives to care for
children who are outside the child welfare system and the purview
of the courts.

Since at least the 1980s, some portion of foster children in this
country have been placed with relatives. Some studies contend that
the increase in the number of foster children being placed with
relatives may have been, at least initially, the result of a
shortage of traditional foster homes. 2 Others suggest that
kinship care increased as a result of the Adoption Assistance and
Child Welfare Act of 1980. This act required states to place
children in the least restrictive (most family like) setting
available, a requirement that has been interpreted by many states
as implying a preference for placing foster children with their
relatives. The increase in kinship care may also stem in part from
litigation (Matter of Eugene F. v. Gross, Sup. Ct., NY County,
Index No. 1125/ 86) that resulted in New York City's bringing
certain children being cared for by relatives into the formal
foster care system and making them eligible for publicly funded
services. Regardless of the historical impetus behind the growth
in kinship care, section 505 of the Personal Responsibility and
Work Opportunity Reconciliation Act of 1996 amended federal law to
require that the states consider giving priority to relatives when
deciding with whom to place children while they are in the foster
care system.

1 The proportion of all foster children nationwide who were
eligible for federal title IV- E funds increased from about 40
percent in 1985 to about 50 percent in 1997. 2 The foster care
population nationwide increased from 280,000 to 400,000 between
1986 and 1990.

GAO/HEHS-99-32 Kinship Foster Care Page 4

B-279199

Kinship care cases are eligible for federal title IV- E funding
if, in addition to other criteria, the caregivers meet state
licensing requirements for foster homes and the child's parents
meet the income eligibility criteria. 3 In 1996, in about 60
percent of the kinship care cases in California and about 50
percent of such cases in Illinois, the caregiver received title
IV- E funding. In the remaining kinship care cases in these
states, the caregiver may have received an Aid to Families with
Dependent Children (AFDC) grant, which may have been a child- only
grant. 4

Thirty- nine states reported in a 1996 survey conducted by the
Child Welfare League of America (CWLA) that in 1995 they had a
total of about 107,000 foster children in kinship care, or about
one- quarter of all foster children in the United States. 5 In
1995, the proportion of all foster children in each state who were
in kinship care ranged from 0.4 to 52 percent. As time passes,
states appear to be relying more on kinship care. CWLA has
reported that between 1990 and 1995, the number of children in
foster care increased by 21 percent (from 400,398 in 1990 to
483,629 in 1995), while the number of kinship care children
increased by 29 percent.

In 1995, the foster care population in California was 87,010, or
about 27 percent larger than it had been in 1990, while the
kinship care population was about 36 percent larger. According to
our survey, as of September 15, 1997, 51 percent of the 74,133
foster children in California who had been in the system since at
least March 1, 1997, were in kinship care. 6

In 1997, the foster care population in Illinois was 50,721, or
about 159 percent larger than it had been in 1990, while the
kinship care population was about 250 percent larger. Up until
July 1995, children whose parents were absent and who were living
safely with a relative were considered neglected under Illinois
state law, and the state generally

3 Miller v. Youakim, 440 U. S. 125 (1979). 4 The Personal
Responsibility and Work Opportunity Reconciliation Act of 1996
replaced AFDC with the TANF block grant program. 5 The survey is
reported in Michale R. Petit and Patrick A. Curtis, Child Abuse
and Neglect: A Look at the States, 1997 CWLA Stat Book
(Washington, D. C.: CWLA Press, 1997). CWLA comprises 900 public
and private agencies across the country that provide a wide array
of services, including child protective services, family
preservation, adoption, and family foster care. Some states do not
have formal foster care because they promote the obligation of
relatives to care for children within the private sphere of the
family, thereby diverting children from the foster care system.
See J. D. Berrick, When Children Cannot Remain Home: Foster Family
Care and Kinship Care, The Future of Children, Vol. 8, No. 1
(Spring 1998), pp. 72- 87.

6 See appendix V, table V. 2, for characteristics of the child and
the foster care setting in foster care cases in California as of
September 15, 1997.

GAO/HEHS-99-32 Kinship Foster Care Page 5

B-279199

assumed custody of such children. In these cases, the relative's
home at the time was frequently converted into kinship care within
the foster care system. This may have accounted for the growth of
the kinship care population in Illinois up until that time.
Illinois amended the definition of neglected child, effective July
1, 1995, and as a result, such children are no longer considered
neglected and the state no longer assumes custody. 7 According to
our survey, as of September 15, 1997, 55 percent of the 48,745
foster children in Illinois who had been in the system since at
least March 1, 1997, were in kinship care. 8

Federal foster care statutes and regulations, which emphasize the
importance of both reunifying families and achieving permanency
for children in a timely manner, apply to all foster care cases,
whether a child is in kinship care or another foster care setting.
Outcomes in foster care cases include (1) family reunification,
(2) adoption, (3) legal guardianship, and (4) independent living
or aging out of the foster care system, usually at age 18. In
emphasizing the goal of family reunification, for example, federal
law requires that the states make reasonable efforts to reunify
foster children with their parents. The law requires that the
states develop case plans that among other things describe the
services that are to be provided to help parents, children, and
foster parents facilitate the children's return to their own safe
home or their permanent placement elsewhere. The states are
required to review foster care cases at least every 6 months and
must hold permanency planning hearings at least every 12 months,
during which a judge or a hearing officer determines whether a
state should continue to pursue the current goal or begin to
pursue some other permanency goal. When foster children cannot be
safely returned to their parents in a timely manner, the Adoption
and Safe Families Act of 1997 (enacted after the period covered by
our survey) includes a provision requiring the states to begin the
process to file a petition to terminate parental rights if a child
has been in foster care for 15 of the most recent 22 months,
unless (1) required reasonable efforts and services to reunify the
family have not been made in accordance with the case plan, (2) a
compelling reason is documented in the case plan indicating why it
would not be in the best interest of the child to terminate
parental rights at that time, or (3) at the option of the state,
the child is being cared for by a relative. At the same time that
the states are required to initiate termination procedures, they
must also identify and recruit

7 Although this change has sharply curtailed growth in the foster
care population in Illinois, this reform was not retroactive.
Therefore, many children remain in foster care in Illinois even
though there is no evidence that they have been neglected or
abused as defined currently.

8 See appendix V, table V. 2, for the characteristics of children
and the foster care setting in foster care cases in Illinois as of
September 15, 1997.

GAO/HEHS-99-32 Kinship Foster Care Page 6

B-279199

qualified families for adoption. Thus, if none of the exceptions
apply, the law attempts to achieve permanency through adoption.

Most research on the quality of kinship care has used the
demographic characteristics of the caregivers as indirect
indicators of the quality of foster care they provide. Although
the studies' results have varied somewhat, many studies have found
that kinship caregivers tend to be older, have less formal
education and lower incomes, are less often married, and are less
healthy than other foster caregivers. 9 On the basis of these
characteristics, child welfare researchers and practitioners have
inferred that the quality of kinship care may be lower than the
quality of care in other foster care settings.

For Most Measurements of Quality, Kinship Care and Other Foster
Care Were Comparable but Some Safety and Quality Assurance
Concerns Remain

Our analysis of the caseworkers' responses to our survey of open
foster care cases in California and Illinois showed that, overall,
the quality of both kinship care and other foster care was good
and that in most respects the experiences of children in kinship
care and in other foster care settings were comparable. In both
states, most caregivers in kinship as well as foster care settings
received high scores from their caseworkers when it came to
performing parenting tasks. We also found that, in general,
children in kinship care in these states experienced significantly
more continuity in their lives that is, continued contact with
family, friends, and the neighborhood they lived in before
entering foster care than other foster children. However, we also
found that while the caseworker in most kinship as well as other
foster care cases believed that the caregivers were likely to
enforce court- ordered restrictions on parental visits, the
proportion of cases in which this view was held was smaller for
kinship care cases than other foster care cases. Moreover,
requirements such as standards or approval criteria for becoming a
caregiver and training for caregivers were less stringent for
kinship care in California and Illinois than for other foster
care.

Caregivers in Kinship Care and Other Settings Performed Parenting
Tasks Adequately or Very Adequately

In both California and Illinois, most kinship and other foster
caregivers received comparably high scores from their caseworker
in performing nearly all the parenting tasks we asked about in our
survey. These tasks covered three areas: (1) providing day- to-
day care, such as providing supervision and emotional support to a
child, setting and enforcing limits on the child's behavior, and
making sure the child attends school;

9 See app. II and app. III, tables III. 8 through III. 13, for
research results regarding the demographics of caregivers.

GAO/HEHS-99-32 Kinship Foster Care Page 7

B-279199

(2) ensuring that the child is up- to- date on routine medical
examinations; and (3) interacting with medical, mental health, and
educational professionals. 10 We found no research that directly
measured foster parents' ability to perform such tasks.

For nearly all the parenting tasks we asked about, the caseworkers
in more than 90 percent of kinship care and other foster care
cases in the two states responded that the caregivers performed
those tasks either adequately or very adequately. A smaller
percentage about 80 percent of the children in kinship care in
Illinois, however, were up- to- date on their routine vision and
dental examinations, compared with 90 percent of other foster
children. 11 State officials in Illinois speculated that this was
because kinship caregivers are more likely than other foster
caregivers to seek vision and dental care for their foster
children only as often as they do for themselves, which is less
frequently than state standards and guidelines call for. Those
officials believed that other foster caregivers are more likely to
follow state standards and guidelines when it comes to their
foster children.

Foster Children in Kinship Care Had More Continuity in Their Lives

In both California and Illinois, responses to our survey questions
indicated that there was significantly more continuity in the
lives of children in kinship care than in other foster care
settings. While many mental health professionals agree that
continuity in relationships is good for children in general, there
is less agreement about the merits of continuity in the lives of
abused or neglected children. Experts do agree that contact with
siblings, and especially living with siblings, is beneficial for a
child and that parental visits with foster children are needed to
achieve reunification when this is an appropriate goal. Experts
also report that a child's familiarity with the caregiver lessens
the trauma of separation from the family, at least in the short
run. Advocates of kinship care further assert that placing a
foster child with relatives or friends may help maintain
continuity in the child's life by maintaining ties with the
child's community, school, and church. Many believe, however, that
parents who neglect or abuse their children learn this behavior
from members of a dysfunctional immediate or extended family. 12
So, living with relatives and continued contact with the community
may not be in the best interest of

10 For a complete list of the parenting tasks we asked about in
our survey, see questions 17, 18, and 19 in the questionnaire in
app. IV. 11 See app. V, table V. 5, for survey results regarding
caregivers' performance of different parenting tasks. 12
Summarized by Berrick in When Children Cannot Remain Home.

GAO/HEHS-99-32 Kinship Foster Care Page 8

B-279199

the child because the child continues to live in the environment
that may have led to the abuse or neglect.

Our survey asked for information about three types of continuity
in foster children's lives: (1) their previous familiarity with
the person who became their foster parent; (2) their contact while
in foster care with their parents, other relatives, and friends;
and (3) their involvement, while in foster care, with the
community they lived in before they entered the system. Our
analysis showed that there was significantly more continuity in
the lives of children in kinship care than in other foster care
settings with respect to nearly all the indicators we used to
measure these three categories of continuity. 13 In general, our
findings were consistent with the results of other research about
the relationship of kinship care and continuity in foster
children's lives. 14

In measuring children's familiarity with the persons who became
their foster parents, the results of our survey in both California
and Illinois indicated that a significantly larger proportion of
children in kinship care than other foster care knew their
caregivers before entering the system. In addition, a
significantly larger proportion of kinship care children had
resided with their caregivers previously. (See fig. 1.)

13 See app. V, table V. 6, for additional survey results regarding
continuity. 14 See app. II and app. III, tables III. 1 through
III. 6, for research results regarding continuity.

GAO/HEHS-99-32 Kinship Foster Care Page 9

B-279199

Figure 1: Children's Familiarity With Their Caregivers in
California and Illinois

In measuring the extent to which foster children were in contact
with their parents, other relatives, and friends in California and
Illinois, in significantly more kinship care than other foster
cases the caseworkers reported that the children were in contact
with family and friends. For example, the caseworkers' responses
to our survey showed that mothers with children in kinship care
(24 percent in California, 39 percent in Illinois) visited their
children more often than specified in their case plans than did
mothers with children in other foster care settings (6 percent in

GAO/HEHS-99-32 Kinship Foster Care Page 10

B-279199

California, 11 percent in Illinois). 15 To put this into
perspective, however, in both kinship care and other foster care
settings, less than 50 percent of mothers visited their children
as often as specified in their case plans. Other research has also
shown that parents of children in kinship care are more likely to
visit their children at least once a year, and visit them more
often per year, than parents of other foster children. 16 In both
California and Illinois, in a significantly larger proportion of
kinship than other foster care cases the caseworkers noted that
one or more of a child's siblings were living in the same foster
home. According to our survey, children in kinship care also had
more contact with their friends and relatives other than parents,
foster parents, or siblings. (See fig. 2.) Other studies reported
similar findings. For example, surveys of foster children in
Baltimore County, Maryland, in 1993 and in California from 1988
through 1991 have shown that children in kinship care were more
likely to live with siblings than were other foster children. 17

15 Caseworkers develop a case plan for each case that indicates
the actions each parent is to take in order to be reunified with a
child, including the level of visitation required or allowed. The
visitation provision is updated periodically as permanency goals
and other circumstances in the case change. When family
reunification is the permanency goal, the plan usually calls for
parents to visit children frequently to build or maintain a
relationship that will allow them to be reunified. However, when
family reunification is no longer deemed possible, parental visits
may still be allowed and specified in the case plan as long as the
child benefits from such contact.

16 Nicole S. Le Prohn, Relative Foster Parents: Role Perceptions,
Motivation and Agency Satisfaction, Ph. D. dissertation,
University of Washington, Seattle, Washington, 1993. (See app. II
and app. III, table III. 3, for research results regarding
parental visits.)

17 Le Prohn, Relative Foster Parents, and J. D. Berrick, R. P.
Barth, and B. Needell, A Comparison of Kinship Foster Homes and
Foster Family Homes: Implications for Kinship Foster Care as
Family Preservation, Children and Youth Services Review, Vol. 16,
Nos. 1- 2 (1994), pp. 33- 63. (See app. II and app. III, tables
III. 4 and III. 5, for research results regarding contact with
siblings.)

GAO/HEHS-99-32 Kinship Foster Care Page 11

B-279199

Figure 2: Children's Contact With Family Members and Friends in
California and Illinois

a The difference between kinship care and other foster care
placements is not statistically significant.

Finally, in measuring children's contact with the communities they
lived in before they entered the system, in significantly more
kinship care than other foster care cases in California and
Illinois caseworkers indicated that children had contact with
their established community. More specifically, in both California
and Illinois a larger proportion of children

GAO/HEHS-99-32 Kinship Foster Care Page 12

B-279199

in kinship care than in other foster care settings lived in the
same neighborhood they had lived in before entering foster care.
18 (See fig. 3.) This is consistent with other studies of foster
children in Illinois. 19 Furthermore, according to our survey, a
larger proportion of children in kinship care in each state were
attending the school they would have attended had they not entered
the system.

18 One recent study showed that the neighborhoods in which
children in kinship care lived were more often considered to be
dangerous than the neighborhoods in which other foster children
lived. This study did not indicate whether the neighborhoods were
the same ones the children lived in before entering foster care.
Specifically, the study noted that while kinship caregivers and
other caregivers perceived their neighborhoods to be good in terms
of quality and safety, a larger proportion of kinship care homes
(22 percent) than other foster care homes (6 percent) were judged
by the person interviewing the caregivers to be located in
dangerous areas. J. D. Berrick and others, Assessment, Support,
and Training for Kinship Care and Foster Care: An Empirically-
Based Curriculum (Berkeley, Calif.: University of California,
Berkeley, Child Welfare Research Center, 1998).

19 See app. II and app. III, table III. 1, for research results
regarding foster children living in the same neighborhoods they
lived in before entering foster care.

GAO/HEHS-99-32 Kinship Foster Care Page 13

B-279199

Figure 3: Children's Contact With the Communities They Lived in
Before Entering Foster Care in California and Illinois

The number of times caregivers changed during a foster care
episode has also been used as an indication of continuity in a
child's life. Previous research in California has shown that
foster caregivers changed fewer times per foster care episode in
kinship care than other foster care cases; the lives of children
in kinship care tended to be more stable while they were in foster
care. 20

20 Berrick, Barth, and Needell, A Comparison of Kinship Foster
Homes. (See app. II and app. III, table III. 6, for research
results regarding number of placements in foster care.)

GAO/HEHS-99-32 Kinship Foster Care Page 14

B-279199

Kinship Caregivers Were Somewhat Less Likely to Enforce
Restrictions on Parental Visits Than Other Foster Caregivers

Our survey suggests that the safety of a somewhat larger
proportion of children in kinship care than other foster care in
California and Illinois may be at risk because their caregivers
may be unwilling to enforce court- ordered restrictions on
parental visits. Specifically, in 72 percent of the California
kinship care cases and 68 percent of the Illinois kinship care
cases in which the parents' visits with their children were
restricted, the caseworkers believed that the caregivers were
likely to take the necessary action to enforce the restrictions.
In contrast, 92 percent of the caseworkers in other foster care
cases in California and 80 percent in other foster care cases in
Illinois believed that the caregivers were likely to enforce
parental visitation restrictions. 21 (See fig. 4.) As noted
earlier, parental visits provide stability for children while they
are in foster care. In some cases, however, the court may restrict
visits by the parents because it believes the child might be
harmed by these visits. 22 In more than 85 percent of our survey
cases, the court had restricted visits by the parents. 23

21 State child welfare officials in California and Illinois
believed that this information alone is not adequate to draw
conclusions about the safety of children placed in kinship care.
To do so, information about the extent to which caregivers allowed
parents to violate the restrictions in these cases, and instances
in which the children had actually been harmed as a result, would
be needed.

22 There are a number of reasons why parental visits are
restricted in foster care cases. Visits may be prohibited when a
parent appears to be under the influence of alcohol or drugs or
may be prohibited unless they are supervised by a caseworker or
another professional. In extreme cases, they may be prohibited
under any circumstances.

23 See app. V, table V. 7, for more information regarding the
caregivers' willingness to enforce parents' visitation
restrictions.

GAO/HEHS-99-32 Kinship Foster Care Page 15

B-279199

Figure 4: Caregivers' Willingness to Enforce Parental Visitation
Restrictions in California and Illinois

Some Quality Assurance Standards Are Lower for Kinship Care Than
Other Foster Care

Certain elements of California's and Illinois's quality assurance
systems are less rigorous for kinship care than for other foster
care settings. Both California and Illinois have less stringent
requirements for becoming a caregiver and provide less training
and support to kinship caregivers. States sometimes treat kinship
caregivers differently because of the family bond that is assumed
to be present between children and their relatives. They believe
this bond mitigates the need for more intrusive state oversight in
these cases. While some experts in child welfare believe that this
exception for kinship caregivers is reasonable, others believe
that while a state has custody of a child, all caregivers should
be held to the same standards.

States Apply Less Stringent Requirements for Kinship Caregivers

To become foster caregivers in California or Illinois, a child's
relatives must meet certain criteria specifically designed for
kinship care that are

GAO/HEHS-99-32 Kinship Foster Care Page 16

B-279199

less stringent than the licensing requirements that apply to other
foster caregivers. For example, since Illinois does not require
kinship caregivers to be licensed, they do not have to meet
licensing requirements regarding the number of bedrooms or the
square footage in the home. Furthermore, they are exempt from some
specific requirements designed to ensure a foster child's safety
in the home.

Even though kinship caregivers are not required to meet the same
requirements as other caregivers, in California if a foster child
is eligible for title IV- E funds, the kinship caregivers receive
the same maintenance payment as licensed caregivers would. Unlike
in California, kinship caregivers in Illinois can receive the same
maintenance payment as other caregivers only if they choose to
meet the licensing requirements of other foster caregivers and
thereby become licensed. Otherwise, relatives must meet less
stringent requirements to provide foster care, which results in a
lower maintenance payment. State child welfare officials in
Illinois indicated that about 50 percent of the kinship caregivers
in the state are licensed to provide foster care. 24

States Require the Same Minimum Number of Caseworker Visits for
Kinship Care and Other Foster Care Cases

Both California and Illinois require caseworkers to periodically
visit all foster children. Caseworkers are required to visit
foster children in order to, among other things, monitor the
quality of the care they are receiving and determine whether the
children or caregivers have any unmet service needs. Generally, in
California, caseworkers are required to visit foster children at
least once a month. When the goal is something other than family
reunification, caseworkers are required to visit at least once
every 6 months, because in these cases the children are considered
to be in a more stable setting. Illinois requires caseworkers to
visit foster children at least once a month, regardless of the
permanency goal.

According to our survey, caseworkers in California and Illinois
visited both foster children in kinship care and those in other
settings more often on average than formally required, but they
visited children in kinship care less often on average than
children in other foster care settings. Eighty- five percent of
our cases in California were past family reunification so were
required to be visited once every 6 months. In California,
caseworkers visited kinship care children an average of 3.8 times
in 6 months compared with an average of 5.3 visits to other foster
children. Similarly, in Illinois caseworkers visited kinship care
children an average of 8 times in 6

24 See app. III, tables III. 14- III. 16, for research results
regarding training, support services, and caseworker visits.

GAO/HEHS-99-32 Kinship Foster Care Page 17

B-279199

months compared with an average of 11.3 visits to other foster
children. 25 Our survey results were consistent with other
research that has also found that caseworkers tend to visit
children in kinship care less frequently than other foster
children. 26

States Provide Training and Support Services to Fewer Kinship
Caregivers Than Other Foster Caregivers

California and Illinois provide fewer kinship caregivers with
training than other foster caregivers. To help ensure good quality
foster care, both states require licensed foster caregivers to
receive training in topics such as the child welfare system and
procedures and caring for children who have been abused or
neglected. Since kinship caregivers are not required to be
licensed in either California or Illinois, a smaller proportion of
kinship caregivers than other foster caregivers in these states
receive such training. Because of funding constraints, California
has historically precluded kinship caregivers from receiving such
training unless they pay for it themselves. Nonetheless,
California state officials believe that kinship caregivers should
receive training that is specifically designed for them. The Child
Welfare Research Center (CWRC) has found that both kinship
caregivers and other foster caregivers in California would like
more training on subjects such as foster parent licensing,
prenatal drug exposure, and how to interact more effectively with
social service agencies. 27 CWRC has also found that kinship
caregivers in California want more information about court
proceedings related to foster care and how to navigate the child
welfare system in order to receive needed services.

Some states provide fewer kinship caregivers with support services
than other foster caregivers. Services such as respite care,
housing support, counseling, transportation, child care, legal
services, and access to support groups are designed to help foster
caregivers successfully perform their role. Research conducted in
California found that a smaller proportion of kinship caregivers
received such services than other foster caregivers. 28 This
research also found that kinship caregivers in California,
reacting to the emotional demands of caring for an abused or
neglected relative, also

25 Child welfare officials in both states did not see a problem
with the difference in average number of visits to kinship care
children and to other foster children because caseworkers were
visiting both types of children at least as often on average as
required. They indicated that additional visits are made when a
caseworker believes they are needed. Illinois officials stated
that the difference in the number of caseworker visits by setting
might reflect caseworkers' attitudes but is not Illinois policy.

26 See app. III, tables III. 15 and III. 16, and app. V, table V.
4, for additional information on survey and other research results
regarding caseworker visits, and Alfreda P. Iglehart, Kinship
Foster Care: Placement Service and Outcome Issues, Children and
Youth Services Review, Vol. 16, Nos. 1- 2 (1994), pp. 107- 22.

27 CWRC is associated with the School of Social Welfare at the
University of California at Berkeley. 28 Le Prohn, Relative Foster
Parents. (See app. II and app. III, table III. 15, for additional
research results regarding services received by caregivers.)

GAO/HEHS-99-32 Kinship Foster Care Page 18

B-279199

wanted to know more about community resources and mental health
services that were available to them. 29

Kinship Care Cases in California Differed From Those in Illinois
With Regard to Permanency Goals and Time in Foster Care

Previous research on children who have left the foster care system
has shown that children who had been in kinship care were less
likely to be adopted and stayed longer in foster care than other
foster children. However, we found no consistent pattern between
California and Illinois. In California, we found a pattern similar
to the research regarding permanency goals among foster care cases
in which a child is still in the system. Specifically, kinship
care cases in California less often had the goal of adoption or
guardianship (and more often had the goal of long- term foster
care) than did other foster care cases. 30 In California, there
was no difference between kinship care and other foster care in
the length of time children spent in foster care. However, in
Illinois, in foster care cases in which a child was still in the
system, a larger proportion of kinship care than other foster care
cases had the goal of adoption and guardianship, and kinship care
cases had been in the system a shorter, not longer, period of
time. Because outcomes for kinship care cases differed in these
two states, it is likely that state foster care policies and
practices rather than the type of foster care setting in which
children were placed had the greatest influence over a foster
child's permanency goal and length of time in care. It should also
be noted that, in both states, we found that most children,
regardless of foster care setting, had been in the system much
longer than they should have been if the Adoption and Safe
Families Act had been in effect at the time of our survey. 31

Research Has Shown That Kinship Care Is Less Likely to End in
Adoption and Length of Stay Is Longer

Several research studies have looked at foster care outcomes and
length of stay. Many of these examined the experiences of a group
of children who entered the system in the same year. Most have
shown that children in kinship care were less likely than other
foster children to be adopted. Most

29 J. D. Berrick and others, Assessment, Support, and Training for
Kinship Care and Foster Care. (See app. II and app. III, table
III. 15, for additional research results regarding services
received by caregivers.)

30 Although state child welfare agencies use the category long-
term foster care in their administrative paperwork to indicate a
potential permanency outcome, they do not consider long- term
foster care a permanency goal, per se, that they would work
toward. Foster children are placed in this category when efforts
to find a home for them outside the foster care system fail.
Recent federal legislation recognizes long- term foster care as a
potential permanency outcome in foster care cases but authorizes
it only when adoption or guardianship is not feasible.

31 This act allows a substantial implementation period and
provides a number of exemptions to the general rule limiting
foster care to 15 months before the state is required to initiate
procedures to terminate parental rights.

GAO/HEHS-99-32 Kinship Foster Care Page 19

B-279199

have also shown that children in kinship care spent more time than
other foster children in the foster care system. 32

Kinship Care Cases in California More Often Had the Goal of Long-
Term Foster Care but Were in the System No Longer Than Other
Foster Care Cases

In California, our analysis of the survey data indicated that
kinship care cases in the foster care system as of September 15,
1997, were more likely to have the goal of long- term foster care
than other foster care cases in the system at that time. Where
reunification was no longer considered feasible, our survey showed
that 67 percent of the cases in kinship care had a goal of long-
term foster care compared with 53 percent of cases in other foster
care settings. (See fig. 5.) The large number of children in
kinship care with the goal of long- term foster care is not
surprising given that according to California officials, the state
had only recently begun to offer adoption and guardianship options
specifically designed for a foster child's relatives. Survey
responses confirmed this belief. In 74 percent of kinship care
cases with a goal of long- term foster care, the caseworkers
responded that the primary reason why the children did not have
adoption as the goal was that they were being cared for by
relatives who did not want to adopt and that moving the children
to another home would be detrimental to them. 33

32 See app. II and app. III, tables III. 18 though III. 20, for
research results regarding permanency and length of stay. 33 See
app. V, tables V. 8 though V. 12, for survey results regarding
permanency and length of stay in California.

GAO/HEHS-99-32 Kinship Foster Care Page 20

B-279199

Figure 5: Goals for California Cases in Which Family Reunification
Was Not Considered Feasible

a In kinship care, 53 percent already had a guardian appointed. b
In kinship care, 93 percent were likely to be adopted and 85
percent were in a preadoptive home. c In other foster placements,
72 percent already had a guardian appointed. d In other foster
placements, 81 percent were likely to be adopted and 62 percent
were in a preadoptive home.

State officials in California pointed out several disincentives
for adoption and guardianship in kinship care cases. Certain
benefits for foster children in California, such as special
priority for assistance in schools and financial assistance for
college, are no longer available when they have been adopted. 34
Similarly, title IV- E maintenance payments are not authorized for
children who leave the foster care system because of legal
guardianship. Guardians who are related to a child could receive a
TANF child- only grant on behalf of the child instead of title IV-
E payments, but this grant is much lower than the title IV- E
maintenance payments. In addition, to qualify for a TANF child-
only grant, the guardian would have to provide proof that the
child attends school and receives medical examinations. According
to our survey, more than half of the open kinship care cases in
California with the goal of guardianship had a guardian

34 The title IV- E Adoption Assistance Program benefits are
available to adopted children who have special needs, including
needs stemming from physical or emotional problems. Payments may
not exceed comparable foster care maintenance payments.

GAO/HEHS-99-32 Kinship Foster Care Page 21

B-279199

appointed but remained in the foster care system. This may be
because guardians can receive the foster care maintenance payment,
which is higher than a TANF child- only grant, if the case remains
in the foster care system. 35

While our survey found that, of all foster care children in
California, 11.3 percent of children in kinship care and 19.1
percent of other foster children had adoption as the goal, in
fact, only 2 percent of the children in foster care were adopted
in 1997. Therefore, the state foster care agency has set the goal
of adoption for many more foster children than are likely to be
adopted, given recent experience.

According to our survey in California, as of September 15, 1997,
children in kinship care had been in the system about as long as
those in other foster care settings. 36 A multivariate analysis of
cases in California confirmed that the type of foster care setting
was not associated with the time foster children had spent in the
system. Both children in kinship care and those in other foster
care settings as of September 15, 1997, had already spent more
than 60 months on average in foster care. This is 45 months longer
than the time now allowed under the Adoption and Safe Families Act
before the states are required to file a petition to terminate
parental rights. Furthermore, we estimate that of the 37,881
children in kinship care in California as of September 15, 1997,
who had been in the system since at least March 1, 1997, nearly 82
percent, or 31,025, had been in the system for 17 months or more.
37 Under federal law, however, children in kinship care may be
excluded from the requirement to terminate parental rights once a
child has been in foster care for 15 of the past 22 months. 38

35 See app. V, tables V. 8 through V. 12, for further survey
results regarding permanency and length of time in foster care. 36
Previous research in California has shown that children in kinship
care stay longer than children in other foster care (see app. III,
table III. 19). Differences in the types of cases studied (open
versus closed foster care cases) or the time period studied may
account for the difference between the results of our survey and
the results of other research.

37 The clock for determining the 15- month requirement for
terminating parental rights begins on the date the case was
adjudicated and the child was determined to have been abused or
neglected, or 60 days from the date when custody was removed from
the parents, whichever came first. We based our estimates on the
more conservative 17- month criteria. See appendix I for a
detailed description of how we arrived at our estimates.

38 Similarly, we estimate that 85 percent, or 30,705, of the cases
in other foster care settings as of September 15, 1997, that had
been in the system since at least March 1, 1997, had also been in
foster care 17 months or more. These cases would not be exempt
from the requirement that states petition to terminate parental
rights unless they meet one of the other exemption criteria in
federal law.

GAO/HEHS-99-32 Kinship Foster Care Page 22

B-279199

Open Kinship Care Cases in Illinois More Often Had Goals of
Adoption or Guardianship and Had Been in the System Less Time Than
Other Foster Care Cases

In contrast to our findings in California, data from our survey in
Illinois indicated that children in kinship care as of September
15, 1997, were more likely to have the goal of adoption or
guardianship than other foster children in the system at that
time. 39 Specifically, 66 percent of kinship care cases had the
goal of adoption or guardianship compared with 47 percent of cases
in other foster care settings. 40 (See fig. 6.) According to state
officials, Illinois has found that kinship caregivers, contrary to
popular belief, are willing to adopt, and Illinois is actively
pursuing adoption in these cases.

Figure 6: Goals for Illinois Cases in Which Family Reunification
Was Not Considered Feasible

a In kinship care, 91 percent were likely to be adopted and 94
percent were in preadoptive homes. b In other foster placements,
82 percent were likely to be adopted and 64 percent were in
preadoptive homes.

While our survey found that in Illinois 41.3 percent of children
in kinship care and 37.9 percent of other foster children had
adoption as a goal, in

39 The most recent data provided by Illinois show that children in
kinship care and other foster care were adopted at similar rates
(see app. III, table III. 20). Differences in the types of cases
studied (open versus closed foster care cases) or the time period
studied may account for the difference between the results of our
survey and the results of other research.

40 See app. V, tables V. 8 though V. 12, for additional survey
results regarding permanency and length of time in foster care up
until September 15, 1997.

GAO/HEHS-99-32 Kinship Foster Care Page 23

B-279199

fact, only 4 percent of all foster children were estimated to have
been adopted in 1997. Therefore, as in California, the state
foster care agency has set the goal of adoption for many more
children than are likely to be adopted, given recent experience.
41

Our survey in Illinois indicated that foster children in kinship
care as of September 15, 1997, had spent 43 months, on average, in
the system. Other foster children had been in care for 53 months,
on average, as of that date. 42 A multivariate analysis of cases
in Illinois also indicated that the type of foster care setting
was associated with the time children had already spent in the
system. Children in kinship care had been in the system about 10
fewer months, on average, than other foster children.

Although children in other foster care settings in Illinois had
spent more months in the system, as of September 15, 1997, than
children in kinship care, foster children in general had spent
much more time, on average, in the system as of that date than the
15 months allowed with the enactment of the Adoption and Safe
Families Act before states are required to file a petition to
terminate parental rights. Furthermore, we estimated that of the
26,712 children in kinship care in Illinois as of September 15,
1997, who had been in the system since at least March 1, 1997, 87
percent, or 23,213, had been in the system for 17 months or more.
As we noted earlier, however, the law allows the states to exclude
children in kinship care from the federal requirement to terminate
parental rights in cases in which they have been in care 15 of the
past 22 months. 43

41 Illinois officials pointed out, however, that the number of
adoptions and guardianships in that state climbed from under 2,000
in 1996 to 3,688 in 1997 and 6,610 in 1998. Furthermore, while
most kinship caregivers are choosing adoption, a significant
proportion is choosing private guardianship because they prefer to
leave their customary family relationship unchanged.

42 Research in Illinois shows that children in kinship care were
less likely to exit from foster care than other foster care
children. Therefore, children in kinship care stayed in the system
longer than other foster children (see app. III, table III. 20).
Differences in the types of cases studied (open versus closed
foster care cases) or the time period studied may account for the
difference between the results of our survey and the results of
other research.

43 Similarly, we estimated that 90 percent, or 19,874, of the
cases in other foster care settings, as of September 15, 1997,
that had been in the system since at least March 1, 1997, had also
been in foster care 17 months or more. If these cases had been
subject to the changes made by the Adoption and Safe Families Act,
the requirement to terminate parental rights would have had to be
enforced unless one of the other exemption criteria had been met.

GAO/HEHS-99-32 Kinship Foster Care Page 24

B-279199

States Are Taking Steps to Help Ensure That Kinship Care Meets the
Needs of Foster Children

Since the fall of 1997, both California and Illinois have been
instituting new programs and practices that are designed to (1)
increase the likelihood that permanent living arrangements will be
found for children in kinship care, as well as other foster care
settings, who cannot return to their parents and (2) continue to
ensure that kinship care is of good quality. They are pursuing
efforts to choose the best kinship caregivers by identifying and
locating a larger pool of relatives to draw from when deciding
with whom to place foster children. To help ensure that children
who cannot return to their parents do not remain in the foster
care system indefinitely, California and Illinois recently enacted
laws and are developing programs that encourage kinship caregivers
and other relatives of foster children to provide permanent homes
for them when necessary. Both states also support adoption and
subsidized guardianship for children in kinship care as pathways
out of the foster care system.

New State Initiatives Are Aimed at Ensuring the Good Quality of
Kinship Care

Both California and Illinois have stepped up their efforts to
identify as many of a foster child's relatives as possible before
deciding with whom to place that child. By expanding the pool of
potential foster caregivers, the states hope to help ensure a
foster child is placed with the relative who is capable of
providing good quality foster care in the short term and who is
willing to provide a long- term home if reunification with the
parents is not feasible. Illinois requires that a diligent search
for the parents when a child enters foster care include a search
for other relatives, as well. The state is contracting with a firm
that specializes in identifying and locating relatives and will
conduct such searches routinely in foster care cases statewide.

Since January 1, 1998, courts in California have had the authority
to order the parents of foster children to disclose the names and
residences of all the children's maternal and paternal relatives.
According to California officials, parents before then typically
provided the names of only one or two relatives, usually the ones
with whom they preferred their child to be placed. In addition,
before a foster child is placed with a relative, California now
applies an expanded assessment requiring that (1) a detailed
background check be conducted; (2) the relative's capacity to help
implement the case plan, including family reunification efforts,
be considered; and (3) the relative's ability and willingness to
provide a permanent home for the child also be considered.

GAO/HEHS-99-32 Kinship Foster Care Page 25

B-279199

Recent legislation in California has also created the Kinship
Support Services Program, one of whose objectives is to help
ensure the good quality of kinship care. Services this program
provides include

 case management;  social services referral and intervention aimed
at maintaining the kinship

family unit for example, housing, homemaker services, respite
care, legal services, and day care;  transportation for medical
care and educational and recreational activities;  individual and
group counseling in parent- child relationships and group

conflict;  counseling and referral services aimed at promoting
permanency,

including kinship adoption and guardianship; and  tutoring and
mentoring for the children.

Both States Have Initiated Programs to Encourage Kinship
Caregivers to Provide Permanent Homes for Foster Children

Both California and Illinois are attempting to help ensure that
children in kinship care spend as little time in the foster care
system as possible. Anticipating federal and state legislation
requiring the states to move more quickly to secure permanent
homes for foster children, including those in kinship care, in
1998 the Illinois Department of Children's and Family Services
instituted new policies and programs related to kinship care to
meet this requirement. In California, the move to encourage
relatives to provide permanent homes for foster children began
with the Governor's Adoption Initiative of 1996, which is a 5-
year plan to identify and implement strategies to maximize
adoption opportunities for children in long- term foster care. In
1996, the state held a policy summit on kinship care that found
that current permanency options present significant cultural and
financial barriers to kin to achieve permanency. Following is an
overview of the activities these states are undertaking to take
better advantage of opportunities for permanently placing foster
children with their relatives.

Kinship Adoption On January 1, 1998, California instituted a
kinship adoption program to remove barriers to adoption by current
kinship caregivers and other relatives of foster children. In a
kinship adoption, caregivers and relatives are permitted to enter
into a kinship adoption agreement, a provision that is not typical
in traditional adoptions. This agreement can address visitation
rights for parents and other family members, as well as how
information about a child is to be shared. The law authorizing the
program sets out procedures for the agreement's enforcement,
modification, and termination. Under the terms of kinship
adoption, parents may voluntarily

GAO/HEHS-99-32 Kinship Foster Care Page 26

B-279199

relinquish their parental rights and designate the relative who
will adopt the child, a provision that is also unique to kinship
adoption.

Concurrent Planning Concurrent planning allows for planning for
the ultimate return of foster children to their parents, as well
as another permanency outcome should family reunification prove
infeasible. This process is intended to shorten the length of time
it takes to secure another permanent home for children once the
court decides that they cannot return to their parents. Illinois
has recently begun concurrent planning; it is particularly useful
when parents have previously been unwilling or unable to provide a
safe home for their children or when repeated clinical
interventions have failed.

The Governor's Adoption Initiative in California also supports
concurrent planning because it attempts to ensure that the long-
term interests of foster children are not sacrificed in favor of
their immediate needs or the interests of the foster caregivers.
According to the second progress report on this initiative, some
foster caregivers who do not wish to or are unsuitable to adopt
their foster children are willing to continue to care for them. As
noted earlier, our survey found this in 74 percent of the kinship
care cases in California that, as of September 15, 1997, had the
goal of long- term foster care. In such cases, the court and child
welfare agency are reluctant to place a foster child with another
family that will and can adopt because the child usually has
already lived with the foster caregiver for a substantial period
of time. Concurrent planning is designed to help ensure that these
permanency issues are considered when deciding in what foster care
setting (including kinship care) a child should be placed.
Furthermore, the second progress report states that

A successful concurrent planning program is one in which the
number of children who enter long- term foster care is
significantly reduced (ideally, eliminated), the time the typical
child spends in the system is reduced, virtually all young
children who do not reunify are adopted rather than placed with
legal guardians, the number of children replaced is reduced
significantly, the proportion of relinquishments increases, and
social workers' comfort with the quality of adoptive families
increases.

Subsidized Guardianship HHS has granted both Illinois and
California a 5- year waiver of the restriction the Social Security
Act places on providing title IV- E maintenance payments to legal
guardians. This waiver enables the states to subsidize
guardianships using title IV- E funds, thus eliminating the
financial disincentive for kinship caregivers to become their
foster child's legal guardian. In its first year, the waiver for
California applies only to children 13 years of age or older. In
each subsequent year, the minimum

GAO/HEHS-99-32 Kinship Foster Care Page 27

B-279199

eligibility age increases by 1 year. When the waiver period ends
in 5 years, all children who were covered by the waiver will have
reached the age of 18, so they will no longer require title IV- E
foster care payments. Thus, California will not be responsible for
any further subsidized guardianship payments for these children
once the waiver period has ended. California recently notified HHS
that it would like to delay the implementation of this waiver
until it has fully analyzed recently passed state legislation that
also provides for subsidized guardianship.

Illinois has received a title IV- E waiver from HHS enabling it to
use title IV- E funds for subsidies to kinship caregivers who
agree to assume legal guardianship of their foster children.
Unlike California, Illinois's subsidy is available for children of
any age. Thus, when this 5- year waiver expires, Illinois will
fund the subsidies for children in this program from state
revenues until they reach the age of 18. Although there are no age
limits under Illinois's waiver, to be eligible a child must have
been in foster care for 1 year and must have lived with the
potential guardian for at least 1 year before that guardian can
apply for payments under this waiver.

Kinship Support Services Program California's Kinship Support
Services Program, described earlier, also

provides an incentive for kinship caregivers to adopt or assume
legal guardianship of their foster children by continuing to make
the program's support services available to them after their
foster children leave the system. Thus, these services are
available to relatives, whether or not the child in their care is
under the jurisdiction of the juvenile court or in the child
welfare system.

Kinship Care Program In 1998, California enacted legislation
requiring that a plan be developed for a Kinship Care Program that
will be separate and distinct from the existing foster care
program and will provide services uniquely suited to the needs of
children being cared for by their relatives. The Department of
Social Services is currently developing a plan for a separate
kinship care program.

Kin- GAP Program California also enacted legislation in 1998 that
set up the Kinship Guardianship Assistance Payment program known
as Kin- GAP. According to California officials, the Kin- GAP
program allows children in kinship foster care to leave the foster
care system by having their kinship caregivers become their legal
guardians. This program allows children who have been assessed as
being in a long- term stable home to exit the foster care system.
Until they reach the age of 18, children in this program have
medical coverage and maintenance payments are made for each

GAO/HEHS-99-32 Kinship Foster Care Page 28

B-279199

child. The law limits this payment to no more than 85 percent of
the title IV- E foster care maintenance payment. By July 1, 1999,
the Department of Social Services must determine what the dollar
amount of the payment will be.

Redefining Permanency Goals In order to reaffirm the priority
Illinois places on securing permanent homes for foster children,
it has established new permanency goals. It has eliminated long-
term relative care as a permanency goal. Illinois officials noted
that caseworkers will thus be forced to more actively seek
permanent homes for children in kinship care and thereby prevent
them from remaining indefinitely in the foster care system simply
because they are being cared for by relatives. New permanency
goals include return home within 5 months, return home within a
year, substitute care pending termination of parental rights,
adoption, guardianship, substitute care pending independence, and
substitute care due to the child's disabilities or mental illness.
44

Conclusions Despite a number of concerns expressed by some child
welfare experts about the quality and outcomes of kinship care
(the setting in which about

one- quarter of the nation's foster children are placed), the
results of our survey of foster care cases in California and
Illinois revealed a positive picture but not without some
cautionary notes. Parenting- skill assessments by caseworkers in
kinship care cases were comparable to parenting- skill assessments
by caseworkers in other foster care cases. This was not true for
other dimensions of quality. Information from our survey suggests
some areas where improvements in kinship care may be needed.
Specifically, there may be cause for concern about health and
safety, especially with regard to observance of the need for
routine dental and eye exams, and about potentially unsafe visits
by abusing parents.

While California and Illinois apply less stringent standards or
approval criteria for kinship caregivers, both states are taking
steps to better ensure good quality kinship care. They are raising
standards for kinship caregivers and widening the pool of
potential kinship caregivers to increase the chances of locating
relatives capable of providing good quality care.

Since the ultimate goal for foster children is a safe and
permanent home, the permanency plan in foster care cases is of
paramount concern. Previous research shows that children in
kinship care cases stay longer in

44 Illinois defines substitute care as care in any setting within
the foster care system.

GAO/HEHS-99-32 Kinship Foster Care Page 29

B-279199

the system and are less likely to be adopted. In our survey, in
California children in kinship care stayed in the system as long
as children in other foster care settings and less often had a
goal of adoption or guardianship. In contrast, in Illinois
children in kinship care stayed in the system a shorter period of
time and more often had a goal of adoption or guardianship than
children in other foster care settings. Differences in permanency
goals and time in foster care, therefore, may depend more on state
policies and practices than on foster care setting. Moreover, both
states have taken initiatives either to make homes with relatives
a viable permanency option or to facilitate permanency planning.

Agency Comments We provided a draft of this report to HHS and
state child welfare officials in California and Illinois for their
review. HHS generally agreed with the

report and also described a number of activities of its
Administration for Children and Families that it believes will
help inform both policy and the child welfare field. HHS also
provided technical comments, which we incorporated where
appropriate. HHS's response is in appendix VI.

California did not provide official comments. However, California
child welfare officials provided oral comments, limited to
technical issues related to information about their programs. We
incorporated their comments where appropriate.

Illinois generally agreed with our report. However, state
officials believed that the standards applied to other foster care
cases with respect to (1) frequency of caseworkers' visits, (2)
criteria for becoming a caregiver, and (3) caregivers' willingness
to enforce parental visitation restrictions should not be applied
to kinship care cases. We believe that it is valid to apply the
same standards in both kinship and other foster care cases as far
as the number of caseworker visits and a caregiver's willingness
to enforce restrictions on parental visits are concerned.
Regarding the number of caseworker visits, we applied the
standards that California and Illinois have already set, which in
both states are the same for kinship and other foster care cases.
Protecting a child's safety should be the overriding concern of
both kinship and other foster caregivers. Therefore, when a
restriction is placed on parental visits in the interest of a
child's safety, it seems reasonable to expect kinship caregivers
to be as willing as other foster caregivers to enforce that
restriction. Although we report that the states apply less
stringent requirements for becoming a kinship caregiver, we have
taken no position on whether the criteria for kinship and other

GAO/HEHS-99-32 Kinship Foster Care Page 30

B-279199

foster caregivers should be equal. We have modified the report to
clarify this.

We will send copies of this report to the Secretary of HHS and
program officials in California and Illinois. We will also send
copies to child welfare program directors in all other states and
make copies available to others upon request. Major contributors
to this report are listed in appendix VII. If you or your staff
have any questions, please contact me at (202) 512- 7215 or
Clarita A. Mrena, Assistant Director, at (415) 904- 2245 or Ann T.
Walker, Evaluator- in- Charge, at (415) 904- 2169.

Sincerely yours, Cynthia M. Fagnoni Director, Education,
Workforce, and

Income Security Issues

GAO/HEHS-99-32 Kinship Foster Care Page 31

Contents Letter 1 Appendix I Scope and Methodology

36 Appendix II Annotated Bibliography of Research on Kinship Care
and Other Foster Care

46 Appendix III Results of Research Comparing Kinship Care and
Other Foster Care

53 Appendix IV Our Foster Care Questionnaire

67 Appendix V Survey Results

84 Appendix VI Comments From the Department of Health and Human
Services

111

GAO/HEHS-99-32 Kinship Foster Care Page 32

Contents

Appendix VII Major Contributors to This Report

114 Related GAO Products 116 Tables Table I. 1: Subquestions on
Whether the Foster Care Setting

Affects a Child's Quality of Care 37

Table I. 2: Subquestions on Whether the Foster Care Setting
Affects a Child's Time in the System and Permanency

37 Table I. 3: Initial and Adjusted Population and Sample Sizes
and

Response Rates for Our Survey of Open Foster Care Cases 39

Table I. 4: Summary of the Results of Our Regression Analyses for
Permanency Goal in California and Illinois

42 Table I. 5: Summary of the Results of Our Regression Analyses
for

Length of Time in Foster Care in California and Illinois 44

Table III. 1: Did the Foster Child Remain in the Same Community or
Neighborhood He or She Lived in Before Entering Foster Care?

53 Table III. 2: How Safe Was the Foster Caregiver's Neighborhood?
53 Table III. 3: Did the Foster Child Maintain Contact With
Parents? 54 Table III. 4: Did the Foster Child Live With Siblings
Who Were in

Foster Care? 55

Table III. 5: Did the Foster Child Maintain Contact With Siblings?
55 Table III. 6: How Many Placements in Foster Care Did the Foster

Child Have? 55

Table III. 7: Did the Foster Child Feel That He or She Was Part of
the Foster Family?

57 Table III. 8: What Was the Foster Caregiver's Age? 57 Table
III. 9: What Was the Foster Caregiver's Marital Status? 58 Table
III. 10: What Was the Foster Caregiver's Education? 59 Table III.
11: What Was the Foster Caregiver's Health? 59 Table III. 12: What
Was the Foster Caregiver's Income? 60 Table III. 13: How Safe Was
the Foster Caregiver's Home? 60 Table III. 14: What Training or
Preparation Did the Foster

Caregiver Receive? 61

Table III. 15: To What Extent Did the Foster Caregiver Receive
Services?

61

GAO/HEHS-99-32 Kinship Foster Care Page 33

Contents

Table III. 16: How Often Did the Caseworker Visit the Foster
Child?

62 Table III. 17: What Required Health Services Did the Foster
Child

Receive? 62

Table III. 18: What Permanency Goals Were Pursued in Foster Care
Cases?

62 Table III. 19: How Long Did the Child Stay in Foster Care? 63
Table III. 20: How Long Was the Foster Child in Care Before

Various Outcomes? 65

Table V. 1: Sampling Errors for Percentage Estimates 84 Table V.
2: Characteristics of the Child and the Setting 85 Table V. 3:
Caregiver's Characteristics 87 Table V. 4: Licensing, Caseworkers'
Visits, and Caregiver's

Training 88

Table V. 5: Caregiver's Performance of Parenting Tasks 89 Table V.
6: Continuity 95 Table V. 7: Caregiver's Willingness to Enforce
Parents' Visitation

Restrictions 101

Table V. 8: Permanency Goals 102 Table V. 9: Cases With the Goal
of Reunification 103 Table V. 10: Cases With the Goal of Adoption
105 Table V. 11: Cases With the Goal of Guardianship 107 Table V.
12: Cases With the Goal of Long- Term Foster Care 109

Figures Figure 1: Children's Familiarity With Their Caregivers in
California and Illinois

10 Figure 2: Children's Contact With Family Members and Friends in

California and Illinois 12

Figure 3: Children's Contact With the Communities They Lived in
Before Entering Foster Care in California and Illinois

14 Figure 4: Caregivers' Willingness to Enforce Parental
Visitation

Restrictions in California and Illinois 16

Figure 5: Goals for California Cases in Which Family Reunification
Was Not Considered Feasible

21 Figure 6: Goals for Illinois Cases in Which Family
Reunification

Was Not Considered Feasible 23

GAO/HEHS-99-32 Kinship Foster Care Page 34

Contents Abbreviations

AFDC Aid to Families with Dependent Children CWLA Child Welfare
League of America CWRC Child Welfare Research Center HHS
Department of Health and Human Services TANF Temporary Assistance
for Needy Families

GAO/HEHS-99-32 Kinship Foster Care Page 35

Appendix I Scope and Methodology

This appendix contains a detailed description of our review of
existing research, interviews with child welfare experts, and
survey of open foster care cases in California and Illinois. We
conducted this review from April 1997 to December 1998 in
accordance with generally accepted government auditing standards.

Literature Review In order to determine what research had been
done on kinship care, we conducted a literature search to identify
journal articles, reports,

dissertations, and theses written between the beginning of 1990
and the fall of 1998 that addressed at least one of the following
two research questions: (1) Does the foster care setting affect
the quality of care a child receives? and (2) Does the foster care
setting affect time in the system and permanency for the child? 45

We began our search by reviewing the bibliographies of three major
publications addressing the subject of kinship care: (1) Child
Welfare League of America, Selected References on Kinship Care
1962- 1994; (2) the Transamerica Systems, Inc., 1997 draft Study
of Outcomes for Children Placed in Foster Care with Relatives; and
(3) Child Welfare League of America, Kinship Care: A Natural
Bridge, issued in 1994. We also conducted a computerized search
for articles written about kinship care after 1994, the latest
year covered in two of these bibliographies. To ensure that we
omitted no major articles on kinship care, we sent copies of the
three bibliographies and the results of the computerized search to
child welfare experts both inside and outside GAO for their
review. These experts suggested several additional articles. To
identify recently published articles while drafting the report, we
updated our computerized search and sent our bibliography to two
additional experts outside GAO for their review. As a result of
this process, we identified more than 150 documents for
preliminary review.

We reviewed these documents to determine whether they met our
criteria for inclusion in our study and whether they reported any
findings related to our research questions. We excluded a number
of documents identified in our preliminary review from our final
compilation of the research, most often because they (1) did not
contain any research results, (2) did not describe original
research but instead summarized others' research, (3) did not
differentiate between kinship and other foster care settings, (4)
did not differentiate between children in the child welfare system
and children

45 We chose 1990 as the earliest year for our search because a
substantial number of children were in kinship care (31 percent)
by 1990.

GAO/HEHS-99-32 Kinship Foster Care Page 36

Appendix I Scope and Methodology

being cared for by relatives outside the child welfare system, (5)
did not include new data that had not already been summarized in
another document written in whole or part by the same authors, and
(6) did not address either of our two research questions.

Tables I. 1 and I. 2 list the subquestions we used in the
literature search and the tables in appendix III that show the
research results for each subquestion.

Table I. 1: Subquestions on Whether the Foster Care Setting
Affects a Child's Quality of Care

Subquestion Table of

research results

Does the foster child live with siblings who are in foster care?
III. 4 Does the foster child maintain contact with siblings? III.
5 Does the foster child maintain contact with parents? III. 3 Does
the foster child remain in the same community or neighborhood he
or she lived in before entering foster care? III. 1

Does the foster child feel that he or she is part of the foster
family? III. 7 What is the foster caregiver's age? III. 8 What is
the foster caregiver's marital status? III. 9 What is the foster
caregiver's education? III. 10 What is the foster caregiver's
health? III. 11 What is the foster caregiver's income? III. 12
What training or preparation did the foster caregiver receive?
III. 14 What required health services does the foster child
receive? III. 17 How often does the caseworker visit the foster
child? III. 16 To what extent does the foster caregiver receive
services? III. 15

Table I. 2: Subquestions on Whether the Foster Care Setting
Affects a Child's Time in the System and Permanency Subquestion

Table of research

results

How long did the foster child stay in foster care? III. 19 How
many placements in foster care has the foster child had? III. 6
How long was the foster child in care before adoption, the goal
changed to adoption, the child was placed with an adoptive family,
or the child was freed for adoption?

III. 20 How long was the foster child in care before reunification
with his or her parents? III. 20

What permanency goals are pursued? III. 18

GAO/HEHS-99-32 Kinship Foster Care Page 37

Appendix I Scope and Methodology

Interviews With Child Welfare Experts

To obtain a broader perspective on the issues surrounding kinship
care, we interviewed researchers, public policy advisers,
physicians, attorneys, family court judges, social workers,
adoption caseworkers, and representatives of organizations that
have an interest in foster care or child welfare in general. We
asked for their opinions about the strengths and weaknesses of
kinship care, the quality of kinship care, additional safeguards
needed in the system, if any, and the effect of kinship care on
foster care outcomes. We also interviewed state program officials
to obtain information about kinship care in their state and their
opinions about kinship care in general.

Survey Methodology We surveyed open foster care cases in
California and Illinois to obtain information about the quality of
care that children in kinship care receive

relative to that of foster children in other foster care settings,
as well as information about the effect of kinship care on
permanency goals and the time children spend in foster care.

Survey Design and Limitations

Each state selected a simple random sample of open foster care
cases for our survey, from all cases that were in its foster care
system on June 1, 1997, and had been there continuously since at
least March 1, 1997. Each sample was intended to represent the
entire population of open foster care cases in the state during
that time. The samples allowed us to make statements about the
experiences of the foster children who made up the foster care
population during that time. Because these samples were not drawn
from a population of all children who entered the foster care
system in a state, however, they do not represent the experiences
of all foster children who entered the system. Foster children who
spend a relatively short time in the system may be
underrepresented in our samples, while children who spend more
time in foster care may be overrepresented. Furthermore, while the
survey results based on these samples can be generalized to the
population of open foster care cases during the specified time in
each state, they do not represent the foster care population
nationally or in any other state. The foster care cases in
California and Illinois combined account for about one- quarter of
the entire foster care population nationwide and about half of all
kinship care cases.

After our samples were drawn, we learned that 22 of the sampled
cases from California and 2 from Illinois had not been in foster
care continuously from March 1, 1997, through June 1, 1997, and we
excluded

GAO/HEHS-99-32 Kinship Foster Care Page 38

Appendix I Scope and Methodology

them from our study. We excluded an additional 57 cases in the
California sample and 17 in the Illinois sample because
information provided in the questionnaire indicated that they had
not been in the foster care system continuously from June 1, 1997,
through September 15, 1997 the date in the questionnaire for which
caseworkers were asked to provide information about their cases.
We assumed that, if all the questionnaires for the cases in each
of the initial samples had been returned to us, additional cases
would have fallen into these two categories. We used the
proportions of each of these types of cases among respondents to
estimate how many nonrespondents would have fallen into these two
categories. Thus, we reduced our initial samples by 25 cases in
California and 6 cases in Illinois. We also adjusted each state's
initial population size by the same proportions. The initial and
adjusted population and sample sizes and survey response rates are
shown by state in table I. 3. The adjusted populations are our
best estimates of the number of foster care cases that were in the
system continuously from March 1, 1997, through September 15,
1997.

Table I. 3: Initial and Adjusted Population and Sample Sizes and
Response Rates for Our Survey of Open Foster Care Cases

Initial population a Initial

sample Adjusted sample Adjusted

population b Survey responses

Survey response

rate

California 100,044 401 297 74,133 227 76% Illinois 51,967 401 376
48,745 292 78% a The initial population is the state foster care
population as of June 1, 1997, for children who had been in foster
care since at least March 1, 1997. b The adjusted population is
the number of foster care children who were in the state's system
continuously from March 1, 1997, through September 15, 1997.

Data Collection We designed a mail questionnaire that asked
caseworkers for information, as of September 15, 1997, about the
individual foster care cases they were assigned to. We chose this
date because it fell just before the date the questionnaires were
scheduled to be mailed out, so when caseworkers received the
questionnaire they were likely to still recall the facts in a case
as of September 15, 1997. Our survey objectives were to collect
(1) data not in other research, (2) data more directly related to
and thus a better indication of the quality of foster care than
the information in other research, and (3) some of the same data
as in other research because the foster care population we
surveyed and the time covered by our survey were not the same as
those in other research. Examples of information our questionnaire
collected that we did not find in existing research include

GAO/HEHS-99-32 Kinship Foster Care Page 39

Appendix I Scope and Methodology

 foster children's knowledge of their foster caregivers before
entering foster care;  foster caregivers' history of child abuse
or neglect, domestic violence, or

drug abuse;  foster caregivers' parenting skills;  health services
foster children received; and  the likelihood that foster
caregivers would enforce restrictions on parental

visits and thus protect children from abusing parents. We
pretested the questionnaire with a number of foster care
caseworkers in California and Illinois and revised it on the basis
of the pretest results. 46 We mailed a questionnaire for each case
in our samples to the manager in the office handling that case,
who was instructed to give it to the caseworker assigned to that
case. The caseworker was asked to respond to the questionnaire
with regard to that case. We conducted multiple follow- ups with
office managers and caseworkers, by both mail and telephone,
encouraging them to respond. In addition to using a mail
questionnaire to collect information about foster care cases in
our samples, we received an automated file from each state that
contained administrative data on each sampled case from that
state. The states rely on these data in managing their foster care
programs. We did not evaluate the validity of these databases.

Estimates of Foster Cases Subject to Termination of Parental
Rights Requirements

Our estimates of the number of foster care cases in each state
that would be subject to the requirement in the Adoption and Safe
Families Act of 1997 to file a petition to terminate parental
rights were based on the number of cases in our samples in which a
child had been in foster care for at least 17 months as of
September 15, 1997. We used 17 months, rather than 15 months as
specified in the law, because the clock for determining whether a
case is subject to the termination of parental rights requirement
begins running on the date the child was adjudicated abused or
neglected or 60 days after the date the child was actually removed
from the parents' custody, whichever came first. Since we did not
know the adjudication date of the cases in our surveys, we used 17
months as a conservative estimate of the time the case would be
subject to the requirement.

Analysis of Survey Data Most of the conclusions we drew from this
survey were based on a comparison within each state of survey
responses for cases in kinship care and cases in other foster care
settings. In each state, we placed each case

46 Appendix IV contains a copy of the final questionnaire.

GAO/HEHS-99-32 Kinship Foster Care Page 40

Appendix I Scope and Methodology

in one of these two groups, depending on the caseworker's response
to a question about the type of foster care setting in that case.
We placed cases in the kinship care category only when the
caseworkers responded that the foster children were in settings
that your state classifies as kinship or relative care. We placed
all other cases in the other foster care setting category. About
half the cases fell into the kinship care group in each state. The
other foster care setting category contained cases in settings
such as substitute care, specialized care, institutional care,
group homes, and traditional foster family homes. The results of
these analyses are contained in appendix V.

We examined the relationship between type of setting and other
variables in the questionnaire by generating crosstabular tables
and statistically testing to determine whether any differences
between two variables in a table were significant at the .05
level. We calculated most of the percentage estimates we reported
in the body of this report and in appendix V using as the base the
number of cases for which there was a response to a variable other
than don't know. For analyses that involved a child's date of
entry into foster care, we used the date that was recorded in the
state's administrative data file. Thus, our calculation of the
average length of time our cross- section of foster children in
each state spent in foster care up until September 15, 1997, was
based on administrative rather than survey data.

Multivariate Analyses In addition to using crosstabulations to
identify the relationship, if any, between two variables, we
performed multivariate analyses. These analyses tested for
associations, at the .05 significance level, between foster care
setting that is, kinship care versus other foster care setting and
permanency goal, as well as the time children spent in foster
care, while taking into account other variables namely, a foster
child's age at entry into foster care, gender, and race and the
parents' history of drug or alcohol abuse that might also
influence the permanency goal or time in the system. 47 For our
multivariate analyses of the relationship between foster care
setting and permanency goal, we constructed a permanency goal
variable by ranking long- term foster care, guardianship, and
adoption according to the extent to which each goal allowed
children and their families to be independent of the foster care
system. Long- term foster care was considered least independent
and assigned a value of 0, guardianship more independent and
assigned a value of 1, and adoption most independent and assigned
a value of 2. We used linear

47 Parents who, according to our survey, were required to undergo
treatment for either drug or alcohol abuse were considered to have
a history of drug or alcohol abuse.

GAO/HEHS-99-32 Kinship Foster Care Page 41

Appendix I Scope and Methodology

regression specifically the ordinary least squares method to
examine the relationship between foster care setting and
permanency goal in foster care cases in each state, while taking
into account the influence other variables may have had on a
permanency goal. We found that there was no significant
relationship between a child's race or gender and his or her
permanency goal in either state. Therefore, we excluded race and
gender from the additional multivariate analyses we conducted.

A regression analysis for cases in California indicated that
foster care setting and a child's age at entry into foster care
were both related to permanency goal. Specifically, children in
kinship care in California were more likely to have long- term
foster care as the goal, and children in other settings were more
likely to have guardianship or adoption as the goal. Our analyses
also indicated that children who entered foster care in California
at an early age were more likely than those who entered at a later
age to have guardianship or adoption as the goal.

A regression analysis for cases in Illinois indicated that foster
care setting, child's age at entry into foster care, and having a
parent with a history of drug or alcohol abuse were all related to
permanency goal. Specifically, in Illinois, children in kinship
care and children who had entered foster care at an early age were
more likely to have guardianship or adoption as the goal than
children in other foster care settings. We also found that
children who had a parent with a history of drug or alcohol abuse
were more likely to have the goal of guardianship or adoption than
children who had parents with no history of drug or alcohol abuse.
See table I. 4 for a summary of the results of our regression
analyses related to permanency goals.

Table I. 4: Summary of the Results of Our Regression Analyses for
Permanency Goal in California and Illinois

Variable Beta p value Variation explained (r 2 contributed)

California Age at entry . 33 .00 .10 Kinship care . 18 .02 .03
Parent in drug treatment . 01 .89 .00 Total variation explained (r
2 ) .13 Illinois

Age at entry . 42 .00 .16 Kinship care .19 .00 .04 Parent in drug
treatment . 13 .03 .02 Total variation explained (r 2 ) .22

GAO/HEHS-99-32 Kinship Foster Care Page 42

Appendix I Scope and Methodology

We also performed a regression analysis to determine the
relationship, if any, between foster care setting and time in
foster care, taking into account the influence of permanency goal,
a child's age at entry into foster care, race, gender, and
parents' history of drug or alcohol abuse. We found that there was
no significant relationship between a child's race, gender, or
having a parent with a history of drug or alcohol abuse and time
in foster care in either state. Therefore, we excluded these
variables from the additional multivariate analyses we conducted
regarding time in foster care.

Our regression analysis for cases in California indicated that
there was no relationship between foster care setting and time in
foster care. The goal of adoption and a child's age at entry into
foster care, however, were both related to time in the system.
Specifically, adoption as the goal explained more than 12 percent
of the variation in the length of time children spent in foster
care. Children with adoption as the goal spent 47 fewer months, on
average, in foster care than children with some other goal. A
child's age at entry explained almost 6 percent of the variation
in the length of time spent in foster care. For each additional
year of age, children spent an average of 2.4 fewer months in
foster care.

Among foster care cases in Illinois, we found that both foster
care setting and the goal of adoption were related to the length
of time children spent in foster care. Specifically, kinship care
and adoption explained 3 percent and 1.4 percent of the variation
in the amount of time children spent in foster care, respectively.
Children in kinship care spent about 9 fewer months in foster
care, on average, than children in other foster care settings.
Similarly, children with the goal of adoption spent about 10 fewer
months in the system, on average, than children with some other
goal. See table I. 5 for a summary of the results of our
regression analyses related to the length of time in foster care.

GAO/HEHS-99-32 Kinship Foster Care Page 43

Appendix I Scope and Methodology

Table I. 5: Summary of the Results of Our Regression Analyses for
Length of Time in Foster Care in California and Illinois

Variable Slope (b) Beta p value Variation explained

(r 2 contributed)

California Age at entry 2.4 . 26 .00 .06 Kinship care 8.1 . 10 .15
.01 Adoption a 47.0 . 42 .00 .12 Total variation explained (r 2 )
.19 Illinois

Age at entry . 80 . 12 .07 .013 Kinship care 9.00 . 15 .02 .030
Adoption 10.10 . 17 .01 .014 Total variation explained (r 2 ) .057
a In an earlier regression analysis, we found that both the goals
of long- term foster care and guardianship were not significantly
related to time spent in foster care. Therefore, we excluded these
variables from the regression analyses summarized in this table.

Statistical Precision of Estimates Because the estimates we
reported from our survey were based on

samples of foster care cases, a margin of error or imprecision
surrounds them. This imprecision is usually expressed as a
sampling error at a given confidence level. We calculated sampling
errors for estimates based on our survey at the 95- percent
confidence level.

The sampling errors for percentage estimates we cited in this
report varied but did not exceed plus or minus 15 percentage
points. This means that if we drew 100 independent samples from
each of our populations samples with the same specifications as
those we used in this study in 95 of these samples the actual
value in the population would fall within no more than plus or
minus 15 percentage points of our estimate.

The sampling error for our estimates of the average number of
visits by caseworkers in each state never exceeded plus or minus
1.3 visits. Sampling errors for our estimates of the average
length of time foster children in each state spent in the system
did not exceed plus or minus 8.7 months. Sampling errors for our
estimates of the number of foster care children in each state who
spent 17 months or more in the system did not exceed plus or minus
2,650 children. Finally, in appendix V, the sampling error for
estimates in each state of the (1) average number of a foster
child's siblings never exceeded plus or minus 0.5 siblings, (2)
average age at which a child entered foster care never exceeded
plus or minus 0.84

GAO/HEHS-99-32 Kinship Foster Care Page 44

Appendix I Scope and Methodology

years, and (3) average age of children in foster care never
exceeded plus or minus 0.92 years.

Because of the relatively small number of responses in some of the
tables in appendix V, and the resulting imprecision of any
population estimates that would be based on those responses,
tables in appendix V with fewer than 41 cases present only the
number of sample cases for which each response was given. We made
no population estimates concerning those responses.

GAO/HEHS-99-32 Kinship Foster Care Page 45

Appendix II Annotated Bibliography of Research on Kinship Care and
Other Foster Care

This appendix contains studies we identified that compare kinship
care and other foster care. A brief description of study design
and methodology follows each item. Appendix I describes how we
identified research in this area and our criteria for including a
study in this bibliography. Appendix III contains the results of
analyses from the studies listed here.

Benedict, Mary I., and R. B. White. Factors Associated with Foster
Care Length of Stay. Child Welfare, Vol. 70, No. 1 (1991), pp. 45-
58.

This article contains the results of a longitudinal study of
children in three urban and suburban jurisdictions in Maryland who
entered foster care for the first time between January 1, 1980,
and December 31, 1983. Data were obtained from the case records of
a random sample of 689 of these children and covered a period that
began the month a child entered foster care and ended in June
1986. A number of factors, such as the parents' ability to care
for and raise children and foster care placement with relatives,
were examined to identify any relationship between them and the
amount of time children spent in foster care.

Berrick, J. D., R. P. Barth, and B. Needell. A Comparison of
Kinship Foster Homes and Foster Family Homes: Implications for
Kinship Foster Care as Family Preservation. Children and Youth
Services Review, Vol. 16, Nos. 1- 2 (1994), pp. 33- 63.

The researchers described the characteristics of a two- stage,
random sample of the 88,000 children in foster care in California
between January 1988 and the date when the article was written in
1991. A screening questionnaire was mailed to the foster parents
of each of the 4,234 children in the initial sample. This sample
was split evenly between traditional and relative foster care
placements. For the screening questionnaire, foster parents
responded in 1,178 (28 percent) of the cases sampled. In 600 of
these cases (246 relative foster care placements and 354
traditional foster care placements), the foster parents completed
a second questionnaire by either telephone or mail. If they cared
for more than one foster child, they were asked to answer the
questions for one child older than 2 who had resided in their home
for at least 6 months. They provided information about the child's
physical and mental health, the types of services the child
received, and their own perceptions of the child welfare agency
and caseworkers. Although the gender, age, and ethnicity of
children in the ultimate sample were similar to those of children
in the total population, the researchers acknowledged that there
was no way to determine the representativeness of the sample of
providers.

GAO/HEHS-99-32 Kinship Foster Care Page 46

Appendix II Annotated Bibliography of Research on Kinship Care and
Other Foster Care

Berrick, J. D., and others. Assessment, Support, and Training for
Kinship Care and Foster Care: An Empirically- Based Curriculum.
Berkeley, Calif.: University of California at Berkeley, Child
Welfare Research Center, 1998.

A chapter in this curriculum reported the results of a study in
which a sample of 161 kin and 96 nonkin caregivers living in the
San Francisco Bay Area were interviewed in their homes. The study
compared the two groups of caregivers on demographics, the quality
of the relationship between caregiver and child, home safety,
neighborhood safety, and other factors related to the quality of
care the children received.

Courtney, M. E. Factors Associated with the Reunification of
Foster Children with Their Families. Social Service Review, March
1994, pp. 81- 108.

This study examined the relationship between factors such as a
child's age, type of foster care placement (kinship or
nonkinship), reason for removal, and the probability that the
child would return to his or her parents. The results were based
on statewide administrative data on a random sample of 8,748 of
the approximately 88,000 children who entered the foster care
system in California for the first time between January 1988 and
May 1991. The author cited as study limitations the short time
period covered by the data, the limited amount of data recorded
for each case, and the quality of items recorded in the database.

Gebel, Timothy J. Kinship Care and Non- Relative Family Foster
Care: A Comparison of Caregiver Attributes and Attitudes. Child
Welfare, Vol. 75, No. 1 (1996), pp. 5- 18.

This study compared the demographics, attitudes, and perceptions
of relative and nonrelative foster parents in one urban county in
a southeastern state in 1993. The results were based on responses
to a questionnaire mailed to the foster parents in random samples
of 140 of the 450 relative foster care cases and 140 of the
approximately 300 nonrelative foster care cases in that county at
that time. Foster parents were asked about their attitudes toward
the use of corporal punishment and their perceptions regarding
children in their care, the behavior of these children, and the
support they received from child welfare agencies. Foster parents
in 111 of the traditional placements and 82 of the placements with
relatives responded to the survey.

GAO/HEHS-99-32 Kinship Foster Care Page 47

Appendix II Annotated Bibliography of Research on Kinship Care and
Other Foster Care

Iglehart, Alfreda P. Kinship Foster Care: Placement Service and
Outcome Issues. Children and Youth Services Review, Vol. 16, Nos.
1- 2 (1994), pp. 107- 22.

This article described the results of a study that compared
selected characteristics of adolescents in kinship care to those
of adolescents not in kinship foster care. Between February and
July 1988, caseworkers in Los Angeles County extracted this
information from the case files of all 1,642 children aged 16 or
older who were in foster care during that period. Data for about
990 adolescents 352 in kinship care and 638 in traditional foster
care were analyzed for this study. Among the characteristics
compared were gender, race and ethnicity, reason for removal,
total number of placements, length of time in current placement,
and degree of agency case monitoring.

Le Prohn, Nicole S. Relative Foster Parents: Role Perceptions,
Motivation and Agency Satisfaction. Ph. D. dissertation,
University of Washington, Seattle, Washington, 1993.

This researcher examined the relationship between relative and
nonrelative placement with respect to what foster parents believed
their role to be, what motivated them to become foster parents,
and how satisfied they were with the foster care agency.
Associations between foster placement type and the children's
behavior and amount of contact with their parents were also
examined. The foster families selected for the study were families
in the Casey Family Program, a long- term foster care program with
offices in 13 states for children who are unable to be reunited
with their birth parents and are unlikely to be adopted. Results
were based on a random sample of about 175 nonrelative foster
homes selected from all nonrelative foster homes in the Casey
program in 1992. That group was compared with the entire
population of about 130 relative foster homes in the Casey program
during 1992. Data were collected from foster parents using a mail
questionnaire and a telephone interview. Eighty- two relative
foster homes and 98 nonrelative homes were included in the
analysis.

Le Prohn, Nicole S., and Peter J. Pecora. Research Report Series:
The Casey Foster Parent Study Research Summary. Seattle, Wash.:
Casey Family Program, 1994.

Same description as for Le Prohn dissertation above.

GAO/HEHS-99-32 Kinship Foster Care Page 48

Appendix II Annotated Bibliography of Research on Kinship Care and
Other Foster Care

Magruder, Joseph. Characteristics of Relative and Non- Relative
Adoptions by California Public Adoption Agencies. Children and
Youth Services Review, Vol. 16, Nos. 1- 2 (1994), pp. 123- 31.

The author compared adoptions in California by relatives and
nonrelatives with respect to children's gender, ethnicity, and
time in placement before adoption and the characteristics of the
adoptive parents and their households. Study results were based on
the 3,214 public adoptions that took place during that state's
fiscal year 1992, for which data were available.

Needell, B. Placement Stability and Permanence for Children
Entering Foster Care as Infants. Ph. D. dissertation, University
of California at Berkeley, Berkeley, California, 1996.

A number of samples were drawn for this study from a longitudinal
database containing all cases in the California Foster Care
Information System from 1988 through 1994. The primary sample
consisted of all 43,066 children in California who entered foster
care before their first birthday and between 1988 and 1994.
Analysis examined the types of placement, length of stay, reasons
for infants' reentry into foster care after reunification, and
factors that may have led to an infant's adoption or
reunification.

Needell B., and others. Performance Indicators for Child Welfare
Services in California: 1994. Berkeley, Calif.: University of
California at Berkeley, School of Social Welfare, Child Welfare
Research Center, 1995.

The results of this study were based on a longitudinal database of
233,000 cases in the California Foster Care Information System.
These children were in foster care during 1988 or had entered care
before the beginning of 1995. The percentage of children in
different types of placements who exited the system by
reunification, adoption, guardianship, and emancipation was
reported, as well as the median length of the children's first
stay in foster care by foster care placement type. The authors
also examined the effect of ethnicity, age at time of entry, and
reasons for removal from the home on the relationships between
placement type and foster care outcome and between placement type
and length of stay.

Needell B., and others. Performance Indicators for Child Welfare
Services in California: 1996. Berkeley, Calif.: University of
California at Berkeley, School of Social Welfare, Child Welfare
Research Center, 1997.

GAO/HEHS-99-32 Kinship Foster Care Page 49

Appendix II Annotated Bibliography of Research on Kinship Care and
Other Foster Care

In this study, the longitudinal database used in the 1995 Needell
and others study cited above was expanded to 300,000 children who
were in foster care during 1988 or had entered care before 1997.
The analyses were similar to those in the 1995 study.

Poindexter, Garthia M. Services Utilization by Foster Parents and
Relatives. Master of Social Work thesis, California State
University, Long Beach, California, 1996.

The author reported on the use of social services by relative and
nonrelative foster parents in Los Angeles County based on 40
foster care cases selected at random from the population of
children who entered foster care in that county during 1994. Of
the 40 cases, 22 were relative foster care placements and 18 were
nonrelative foster care placements.

Scannapieco, Maria, Rebecca L. Hegar, and Catherine McAlpine.
Kinship Care and Foster Care: A Comparison of Characteristics and
Outcomes. Families in Societies, Vol. 78, No. 5 (1997), pp. 480-
88.

From case file information for a cross- section of children in
foster care in Baltimore County on March 23, 1993, the researchers
attempted to determine whether there were differences between
kinship and other foster care placements in terms of permanency
planning goals. Of the 106 children sampled, 47 were in kinship
care and 59 were in other types of placements.

Testa, Mark F. Home of Relative (HMR) Program in Illinois Interim
Report. Chicago, Ill.: University of Chicago, School of Social
Service Administration, 1993.

The author used a database that included information about all
children in foster care in Illinois between fiscal years 1965 and
1992 to establish trends in kinship care placements in Illinois
and to describe various characteristics of foster children and
their foster care outcomes.

Testa, Mark F. Kinship Care in Illinois. In J. D. Berrick, R. P.
Barth, and N. Gilbert (eds.), Child Welfare Research Review, Vol.
2. New York: Columbia University Press, 1997. Pp. 101- 29.

Focusing on reunification and discharge rates among children in
foster care in Illinois between fiscal years 1976 and 1992, the
researcher

GAO/HEHS-99-32 Kinship Foster Care Page 50

Appendix II Annotated Bibliography of Research on Kinship Care and
Other Foster Care

examined the effect of selected factors such as age, race, and
type of foster care placement on the likelihood of reunification
or discharge.

Testa, Mark F. Professional Foster Care: A Future Worth Pursuing?
Child Welfare: Special Edition on Family Foster Care in the 21st
Century. Forthcoming.

This study examined the relationship between children's placement
type and whether or not they (1) remained close to their community
of origin, (2) were placed with other siblings in the same
household, and (3) achieved permanency or stayed in the same
foster care setting. The researcher used administrative data from
Cook County, Illinois, for three different foster care recruitment
programs and two random samples, one of 995 kinship care and one
of 852 traditional foster care placements. The samples included
only placements between December 1, 1994, and September 30, 1996.
Administrative data through September 30, 1997, were used to
determine whether or not the children stayed in one foster care
setting or left the foster care system.

Thornton, Jesse L. Permanency Planning for Children in Kinship
Foster Homes. Child Welfare, Vol. 70, No. 5 (1991), pp. 593- 601.

Three surveys were conducted in this study. Semi- structured
interviews were administered to a random sample of 20 kinship
caregivers in New York City to determine their attitudes toward
adoption. Eighty- six foster care caseworkers in New York City
completed questionnaires that asked for their perceptions about
kinship caregivers' willingness to adopt. Finally, to compare
permanency goals for children in kinship care to those for
children in traditional care, the records from 95 active kinship
foster care cases in April 1985 were examined along with
statistics from an administrative database.

U. S. General Accounting Office. Foster Care: Children's
Experiences Linked to Various Factors; Better Data Needed,
GAO/HRD-91-64. Washington, D. C.: Sept. 11, 1991.

Data on children who entered or left foster care in 1986 in
Georgia, Illinois, New York, Oregon, South Carolina, and Texas and
Los Angeles County and New York City were analyzed for the
relationship of age, ethnicity, gender, location, reason for
entry, and foster care placement type to length of stay. For
Georgia, Oregon, South Carolina, and Texas, computerized data
files of the case records for all children entering or leaving
foster care

GAO/HEHS-99-32 Kinship Foster Care Page 51

Appendix II Annotated Bibliography of Research on Kinship Care and
Other Foster Care

during 1986 were used. For New York, Illinois, Los Angeles County,
and New York City, random samples of children who had been
discharged from foster care during 1986 were used; the New York
and Illinois samples each contained 1,488 children, the sample for
Los Angeles County contained 209 children, and the sample for New
York City contained 130 children.

U. S. General Accounting Office. Foster Care: Health Needs of Many
Young Children Are Unknown and Unmet, GAO/HEHS-95-114. Washington,
D. C.: May 26, 1995.

A random sample of 137 case records of foster children who had
been in either kinship or traditional care exclusively was
selected from the case records of all foster children younger than
3 years old in Los Angeles County and New York City during 1991 to
examine the relationship between placement type and the receipt of
health services by foster children in this age group.

U. S. General Accounting Office. Foster Care: Services to Prevent
Out- of- Home Placements Are Limited by Funding Barriers, GAO/HRD-
93-76. Washington, D. C.: June 29, 1993.

In this study of the statutory and fiscal barriers the states
faced in delivering child welfare services, the researchers used
caseload data for the last day of either calendar or fiscal year
1992 in California, Michigan, and New York to describe trends in
foster care and child welfare services.

Wulczyn, F. H., and R. M. George. Foster Care in New York and
Illinois: The Challenge of Rapid Change. Social Service Review,
June 1992, pp. 278- 94.

Aggregated administrative data on all children in New York's child
welfare system and similar data from Illinois were used to compare
child welfare trends in these two states from 1983 through 1989.
Shifts in total caseload size, average age of children entering
foster care, and the number of relative foster care placements
were examined. The researchers also determined the proportion of
children admitted to foster care during 1988 in each state who
were (1) discharged within 12 months, (2) discharged between 12
and 24 months, and (3) still in the system after 24 months. They
compared the proportions in kinship care placements with those in
nonkinship care placements.

GAO/HEHS-99-32 Kinship Foster Care Page 52

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

This appendix contains the results of analyses from the studies we
identified that compared kinship care and other foster care. These
results are presented in tables organized by research question.
Sources are noted after each table. In some instances, the results
in the tables were based on data from entire populations of foster
children. When they were based on data from samples of foster
children, if the researcher reported that a difference between
kinship and other foster care was statistically significant, the
significance level is noted in parentheses in the table. Appendix
II contains a description of the design and methodology of the
studies in this appendix.

Table III. 1: Did the Foster Child Remain in the Same Community or
Neighborhood He or She Lived in Before Entering Foster Care?

Of nonemergency first placements in Chicago, percentage located in
the same community or neighborhood in which the parents or
guardians resided a Kinship care Other foster

care settings

1991 84.0 50.2 1989 82.0 53.0 1987 76.0 59.0 a Mark F. Testa,
Kinship Care in Illinois, in J. D. Berrick, R. P. Barth, and N.
Gilbert (eds.), Child Welfare Research Review, Vol. 2 (New York:
Columbia University Press, 1997), pp. 101- 29.

Table III. 2: How Safe Was the Foster Caregiver's Neighborhood?

Kinship care Other foster care settings

Percentage of cases in which the interviewer thought the foster
caregiver's neighborhood was dangerous a (. 001)

22 6 a J. D. Berrick and others, Assessment, Support, and Training
for Kinship Care and Foster Care: An Empirically- Based Curriculum
(Berkeley, Calif.: University of California at Berkeley, Child
Welfare Research Center, 1998).

GAO/HEHS-99-32 Kinship Foster Care Page 53

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 3: Did the Foster Child Maintain Contact With Parents?

Kinship care Other foster care settings

Average number of times foster child visited parents in the past
year a

Mother (. 001) 34.89 3.82 Father 16.25 2.07 Percentage of children
who had contact with their mothers (. 01) or fathers a

Did not see parents in past 12 months Mother 19.5 41.8 Father 37.8
39.8 Saw parents at least once in the past 12 months

Mother 56.1 37.8 Father 36.6 23.5 Parents' whereabouts were
unknown

Mother 7.3 11.2 Father 18.3 28.6 Parents were deceased

Mother 17.1 9.2 Father 7.3 8.2 Percentage of foster children who
saw their parents b

At least once a month 56 32 More than four times a month (. 01) 19
3 a Nicole S. Le Prohn, Relative Foster Parents: Role Perceptions,
Motivation and Agency Satisfaction, Ph. D. dissertation,
University of Washington, Seattle, Washington, 1993. b J. D.
Berrick, R. P. Barth, and B. Needell, A Comparison of Kinship
Foster Homes and Foster Family Homes: Implications for Kinship
Foster Care as Family Preservation, Children and Youth Services
Review, Vol. 16, Nos. 1- 2 (1994), pp. 33- 63.

GAO/HEHS-99-32 Kinship Foster Care Page 54

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 4: Did the Foster Child Live With Siblings Who Were in
Foster Care?

Kinship care Other foster

care settings

Percentage of foster children placed with siblings also in foster
care a (. 01) 54.2 28.6

Of the foster families with more than one foster child, percentage
in which siblings were placed together b

Four or more siblings (. 05) 19 7 At least two siblings (. 001) 95
52 Percentage of children placed with siblings c 45 41 a Nicole S.
Le Prohn, Relative Foster Parents: Role Perceptions, Motivation
and Agency Satisfaction, Ph. D. dissertation, University of
Washington, Seattle, Washington, 1993. b J. D. Berrick, R. P.
Barth, and B. Needell, A Comparison of Kinship Foster Homes and
Foster Family Homes: Implications for Kinship Foster Care as
Family Preservation, Children and Youth Services Review, Vol. 16,
Nos. 1- 2 (1994), pp. 33- 63.

c Maria Scannapieco, Rebecca L. Hegar, and Catherine McAlpine,
Kinship Care and Foster Care: A Comparison of Characteristics and
Outcomes, Families in Societies, Vol. 78, No. 5 (1997), pp. 480-
88.

Table III. 5: Did the Foster Child Maintain Contact With Siblings?
Kinship

care Other foster care settings

Average number of times foster children visited their siblings in
the past year a (. 001) 90.2 13.8

Percentage of foster children who visited their siblings at least
once a year a 59.7 63.4 a Nicole S. Le Prohn, Relative Foster
Parents: Role Perceptions, Motivation and Agency Satisfaction, Ph.
D. dissertation, University of Washington, Seattle, Washington,
1993.

Table III. 6: How Many Placements in Foster Care Did the Foster
Child Have? Kinship

care Other foster care settings

Average number of placements a In care less than 30 days 0.81 1.76
In care 30 days or more (. 001) 2.42 4.58 Total placements (. 01)
3.24 6.30 Percentage of foster children with b (. 01)

1 placement 49 37 5 or more placements 9 14 Percentage of foster
children who entered care between 1988 and 1990 in California and
had placements within 4 years after entry c

Placed in a family 1 placement 62.3 56.9

(continued)

GAO/HEHS-99-32 Kinship Foster Care Page 55

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Kinship care Other foster

care settings

2 placements 26.2 24.2 3 placements 7.5 10.3 4 or more placements
4.1 8.7 Open cases

1 placement 53.5 28.5 2 placements 24.5 24.6 3 placements 11.6
17.4 4 or more placements 10.4 29.5 Percentage of foster children
who had at least one placement before current placement d (. 001)
22 34

Percentage of infants in foster care who entered care between 1988
and 1990 in California and had a given number of placements who
were still in care 4 years later e

Throughout California excluding Los Angeles 1 placement 38.0 22.1
2 placements 34.1 34.8 3 placements 13.7 19.6 4 placements 8.5
10.3 5 placements 5.6 13.1 Los Angeles

1 placement 59.9 44.6 2 placements 26.3 30.3 3 placements 8.4 14.5
4 placements 3.6 6.6 5 placements 1.9 4.0 a Nicole S. Le Prohn and
Peter J. Pecora, Research Report Series: The Casey Foster Parent
Study Research Summary (Seattle, Wash.: Casey Family Program,
1994). b Alfreda P. Iglehart, Kinship Foster Care: Placement
Service and Outcome Issues, Children and Youth Services Review,
Vol. 16, Nos. 1- 2 (1994), pp. 107- 22. c B. Needell and others,
Performance Indicators for Child Welfare Services in California:
1994 (Berkeley, Calif.: University of California at Berkeley,
School of Social Welfare, Child Welfare Research Center, 1995).

d J. D. Berrick, R. P. Barth, and B. Needell, A Comparison of
Kinship Foster Homes and Foster Family Homes: Implications for
Kinship Foster Care as Family Preservation, Children and Youth
Services Review, Vo. 16, Nos. 1- 2 (1994), pp. 33- 63.

e B. Needell, Placement Stability and Permanence for Children
Entering Foster Care as Infants, Ph. D. dissertation, University
of California at Berkeley, Berkeley, California, 1996.

GAO/HEHS-99-32 Kinship Foster Care Page 56

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 7: Did the Foster Child Feel That He or She Was Part of
the Foster Family?

Percentage of foster children with different degrees of
integration according to foster parents (. 001) and social workers
a (. 001) Kinship

care Other foster care settings

Children who felt that they were very much part of the foster
family Foster parents 79.0 50.0 Social workers 46.3 15.3 Children
who felt somewhat like a foster child

Foster parents 14.8 29.6 Social workers 37.8 25.5 Children who
felt very much like a foster child

Foster parents 6.2 20.4 Social workers 15.9 59.2 a Nicole S. Le
Prohn, Relative Foster Parents: Role Perceptions, Motivation and
Agency Satisfaction, Ph. D. dissertation, University of
Washington, Seattle, Washington, 1993.

Table III. 8: What Was the Foster Caregiver's Age? Kinship

care Other foster care settings

Average age of foster caregivers in years Foster fathers a 50.28
48.05 Foster mothers a (. 01) 50.23 46.26 Male foster caregivers b
(. 05) 50 47 Female foster caregivers b (. 05) 48 46 Percentage of
female foster caregivers 55 years of age or older b (. 01) 29 19

Percentage of primary female foster caregivers by age c (. 005)
Younger than 18 years 1.2 0.0 Between 18 and 25 2.5 1.9 Between 26
and 40 11.1 32.4 Between 41 and 60 64.2 57.4 Older than 60 21.0
8.3 a Nicole S. Le Prohn, Relative Foster Parents: Role
Perceptions, Motivation and Agency Satisfaction, Ph. D.
dissertation, University of Washington, Seattle, Washington, 1993.
b J. D. Berrick, R. P. Barth, and B. Needell, A Comparison of
Kinship Foster Homes and Foster Family Homes: Implications for
Kinship Foster Care as Family Preservation, Children and Youth
Services Review, Vol. 16, Nos. 1- 2 (1994), pp. 33- 63.

c Timothy J. Gebel, Kinship Care and Non- Relative Family Foster
Care: A Comparison of Caregiver Attributes and Attitudes, Child
Welfare, Vol. 75, No. 1 (1996), pp. 5- 18.

GAO/HEHS-99-32 Kinship Foster Care Page 57

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 9: What Was the Foster Caregiver's Marital Status?
Kinship

care Other foster care settings

Percentage of married foster caregivers a (. 05) 64 83 Percentage
of single foster caregivers b (. 001) 52 24 Percentage of married
foster caregivers by gender c

Foster mothers (. 01) 46.25 80.41 Foster fathers 92.50 97.50
Percentage of primary female foster caregivers by marital status d

Widowed 17.3 11.1 Separated 12.3 5.5 Divorced 19.8 16.7 Married
38.3 55.6 Never married 12.3 11.1 a Maria Scannapieco, Rebecca L.
Hegar, and Catherine McAlpine, Kinship Care and Foster Care: A
Comparison of Characteristics and Outcomes, Families in Societies,
Vol. 78, No. 5 (1997), pp. 480- 88.

b J. D. Berrick, R. P. Barth, and B. Needell, A Comparison of
Kinship Foster Homes and Foster Family Homes: Implications for
Kinship Foster Care as Family Preservation, Children and Youth
Services Review, Vol. 16, Nos. 1- 2 (1994), pp. 33- 63.

c Nicole S. Le Prohn, Relative Foster Parents: Role Perceptions,
Motivation and Agency Satisfaction, Ph. D. dissertation,
University of Washington, Seattle, Washington, 1993. d Timothy J.
Gebel, Kinship Care and Non- Relative Family Foster Care: A
Comparison of Caregiver Attributes and Attitudes, Child Welfare,
Vol. 75, No. 1 (1996), pp. 5- 18.

GAO/HEHS-99-32 Kinship Foster Care Page 58

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 10: What Was the Foster Caregiver's Education? Kinship

care Other foster care settings

Percentage of foster caregivers who had completed high school a 87
78

Mean number of years of school completed b Foster fathers (. 05)
12.32 13.74 Foster mothers (. 001) 11.65 14.02 Percentage of
foster caregivers who did not have a high school diploma c

Female (. 001) 26 10 Male (. 01) 20 9 Percentage of primary female
foster caregivers with education by category d (. 00001)

College graduate 2.5 22.2 Some college 21.5 28.7 High school
graduate 25.3 29.6 Some high school 34.2 16.7 8th grade or less
16.5 2.8 a Maria Scannapieco, Rebecca L. Hegar, and Catherine
McAlpine, Kinship Care and Foster Care: A Comparison of
Characteristics and Outcomes, Families in Societies, Vol. 78, No.
5 (1997), pp. 480- 88.

b Nicole S. Le Prohn, Relative Foster Parents: Role Perceptions,
Motivation and Agency Satisfaction, Ph. D. dissertation,
University of Washington, Seattle, Washington, 1993. c J. D.
Berrick, R. P. Barth, and B. Needell, A Comparison of Kinship
Foster Homes and Foster Family Homes: Implications for Kinship
Foster Care as Family Preservation, Children and Youth Services
Review, Vol. 16, Nos. 1- 2 (1994), pp. 33- 63.

d Timothy J. Gebel, Kinship Care and Non- Relative Family Foster
Care: A Comparison of Caregiver Attributes and Attitudes, Child
Welfare, Vol. 75, No. 1 (1996), pp. 5- 18.

Table III. 11: What Was the Foster Caregiver's Health? Percentage
of foster caregivers

in fair or poor health a Kinship care Other foster

care settings

Male (. 001) 25 6 Female (. 001) 20 7 a J. D. Berrick, R. P.
Barth, and B. Needell, A Comparison of Kinship Foster Homes and
Foster Family Homes: Implications for Kinship Foster Care as
Family Preservation, Children and Youth Services Review, Vol. 16,
Nos. 1- 2 (1994), pp. 33- 63.

GAO/HEHS-99-32 Kinship Foster Care Page 59

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 12: What Was the Foster Caregiver's Income? Kinship

care Other foster care settings

Percentage of foster caregivers whose income was less than $15,000
a year a 88 90

Percentage of foster families with income by category b (. 01)
More than $30,000 24.3 56.0 Less than $10,000 33.8 2.4 Average
foster family income c

Average annual gross income, including foster care payments (.
001) $32,424 $51,320

Average annual income, disregarding money received from either Aid
to Families with Dependent Children (AFDC)- Family Grant or AFDC
Foster Care

$21,854 $36,402 Percentage of primary female caregivers with
household income by category d (. 000005)

More than $40,000 5.2 25.0 $30,001-$ 40,000 3.9 10.2 $20,001-$
30,000 9.1 27.8 $10,001-$ 20,000 22.1 26.8 $10,000 or less 59.7
10.2 a Maria Scannapieco, Rebecca L. Hegar, and Catherine
McAlpine, Kinship Care and Foster Care: A Comparison of
Characteristics and Outcomes, Families in Societies, Vol. 78, No.
5 (1997), pp. 480- 88.

b Nicole S. Le Prohn, Relative Foster Parents: Role Perceptions,
Motivation and Agency Satisfaction, Ph. D. dissertation,
University of Washington, Seattle, Washington, 1993. c J. D.
Berrick, R. P. Barth, and B. Needell, A Comparison of Kinship
Foster Homes and Foster Family Homes: Implications for Kinship
Foster Care as Family Preservation, Children and Youth Services
Review, Vol. 16, Nos. 1- 2 (1994), pp. 33- 63.

d Timothy J. Gebel, Kinship Care and Non- Relative Family Foster
Care: A Comparison of Caregiver Attributes and Attitudes, Child
Welfare, Vol. 75, No. 1 (1996), pp. 5- 18.

Table III. 13: How Safe Was the Foster Caregiver's Home?

Percentage of foster caregivers who a Kinship care Other foster

care settings

Had a fire extinguisher (. 001) 65 94 Had a complete first aid kit
(. 001) 58 95 Knew cardiopulmonary resuscitation (. 001) 57 93 a
J. D. Berrick and others, Assessment, Support, and Training for
Kinship Care and Foster Care: An Empirically- Based Curriculum
(Berkeley, Calif.: University of California at Berkeley, Child
Welfare Research Center, 1998).

GAO/HEHS-99-32 Kinship Foster Care Page 60

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 14: What Training or Preparation Did the Foster
Caregiver Receive?

Kinship care Other foster

care settings

Percentage who felt that training adequately prepared them to be a
foster parent a (. 01) 74.3 55.7

Percentage who received training b (. 001) 13 76 a Nicole S. Le
Prohn, Relative Foster Parents: Role Perceptions, Motivation and
Agency Satisfaction, Ph. D. dissertation, University of
Washington, Seattle, Washington, 1993. b J. D. Berrick, R. P.
Barth, and B. Needell, A Comparison of Kinship Foster Homes and
Foster Family Homes: Implications for Kinship Foster Care as
Family Preservation, Children and Youth Services Review, Vol. 16,
Nos. 1- 2 (1994), pp. 33- 63.

Table III. 15: To What Extent Did the Foster Caregiver Receive
Services? Kinship

care Other foster care settings

Percentage of foster caregivers who received services a
Specialized training (. 001) 5 59 Support group (. 001) 15 62
Respite care (. 001) 6 23 Mean number of services foster
caregivers received a (. 001) 0.53 2.30 Percentage who b

Ordered clothing 77.3 100 Used child health disease prevention
services 86.4 55.7 Mean number of social worker visits with a
foster parent in the past 12 months c 25.13 19.46

Mean number of social worker telephone contacts with a foster
parent in the past 12 months c 37.38 32.78 a J. D. Berrick, R. P.
Barth, and B. Needell, A Comparison of Kinship Foster Homes and
Foster Family Homes: Implications for Kinship Foster Care as
Family Preservation, Children and Youth Services Review, Vol. 16,
Nos. 1- 2 (1994), pp. 33- 63.

b Garthia M. Poindexter, Services Utilization by Foster Parents
and Relatives, Master of Social Work thesis, California State
University, Long Beach, California, 1996. c Nicole S. Le Prohn,
Relative Foster Parents: Role Perceptions, Motivation and Agency
Satisfaction, Ph. D. dissertation, University of Washington,
Seattle, Washington, 1993.

GAO/HEHS-99-32 Kinship Foster Care Page 61

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 16: How Often Did the Caseworker Visit the Foster
Child? Kinship

care Other foster care settings

Percentage of foster children who were not well known to the
caseworker a (. 0001) 33 22

Mean number of caseworkers' visits with foster children during a
6- month period a (. 05) 2.1 3.3

Mean number of caseworkers' visits with foster children in past 12
months b 22.56 20.42

Average number of hours per month foster children spent with a
caseworker c (. 01) 0.65 0.88

Percentage of foster children who had not been visited by a
caseworker in the past month c 46 35 a Alfreda P. Iglehart,
Kinship Foster Care: Placement Service and Outcome Issues,
Children and Youth Services Review, Vol. 16, Nos. 1- 2 (1994), pp.
107- 22. b Nicole S. Le Prohn, Relative Foster Parents: Role
Perceptions, Motivation and Agency Satisfaction, Ph. D.
dissertation, University of Washington, Seattle, Washington, 1993.
c J. D. Berrick, R. P. Barth, and B. Needell, A Comparison of
Kinship Foster Homes and Foster Family Homes: Implications for
Kinship Foster Care as Family Preservation, Children and Youth
Services Review, Vol. 16, Nos. 1- 2 (1994), pp. 33- 63.

Table III. 17: What Required Health Services Did the Foster Child
Receive? Percentage of foster children up to 3 years old who

received health- related services a Kinship care Other foster

care settings

Routine health care services 81.4 93.2 No services (. 10) 16.0 6.8
a U. S. General Acounting Office, Foster Care: Health Needs of
Many Young Children Are Unknown and Unmet, GAO/HEHS-95-114
(Washington, D. C.: May 26, 1995).

Table III. 18: What Permanency Goals Were Pursued in Foster Care
Cases?

Percentage of foster cases by permanency goal a Kinship care Other
foster

care settings

Return to parents 1 14 Adoption 10 38 Independent living 88 42 a
Jesse L. Thornton, Permanency Planning for Children in Kinship
Foster Homes, Child Welfare, Vol. 70, No. 5 (1991), pp. 593- 601.

GAO/HEHS-99-32 Kinship Foster Care Page 62

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Table III. 19: How Long Did the Child Stay in Foster Care? Kinship

care Other foster care settings

Mean number of days a (. 05) 1,008 534 Mean number of months b
93.48 88.02 Average number of years

1994 study (. 001) c 3.3 2.3 1993 study d 3.2 2.8 Percentage of
first admissions in 1988 by length of stay e New York City

12 months or less 5 42 12- 24 months 3 5 Still in care as of June
1990 88 50 Cook County, Illinois

12 months or less 40 50 12- 24 months 10 8 Still in care as of
June 1990 45 40 Percentage of foster children entering care in
1986 in foster care for 1 year or longer f

Georgia 75 52 Oregon 31 40 South Carolina 39 62 Texas 58 47
Percentage of foster children who were in care as of June 30,
1992, by 2- year fiscal period in which they entered f

1977- 78 1.4 0.3 1979- 80 2.6 0.4 1981- 82 3.1 1.1 1983- 84 6.7
3.4 1985- 86 9.7 6.3 Likelihood of being in care for 1 year or
longer (explained in footnote g)

Georgia 2.76 1.00 Oregon 0.67 1.00 South Carolina 0.38 1.00 Texas
1.64 1.00 Kinship care associated with longer length of stay when
controlling for other factors h (. 007) Years not

specified Years not specified Percentage difference between the
likelihood that foster children who entered in a 2- year fiscal
period would be discharged and the likelihood that children in
other foster care settings who entered the system before 1977
would be discharged (explained in footnote i)

1977- 78 10 0 (continued)

GAO/HEHS-99-32 Kinship Foster Care Page 63

Appendix III Results of Research Comparing Kinship Care and Other
Foster Care

Kinship care Other foster

care settings

1979- 80 25 +14 1981- 82 20 +6 1983- 84 10 +2 1985- 86 9 5 1987-
88 37 19 1989- 90 50 24 1991- 92 77 38 a Maria Scannapieco,
Rebecca L. Hegar, and Catherine McAlpine, Kinship Care and Foster
Care: A Comparison of Characteristics and Outcomes, Families in
Societies, Vol. 78, No. 5 (1997), pp. 480- 88.

b Nicole S. Le Prohn, Relative Foster Parents: Role Perceptions,
Motivation and Agency Satisfaction, Ph. D. dissertation,
University of Washington, Seattle, Washington, 1993. c J. D.
Berrick, R. P. Barth, and B. Needell, A Comparison of Kinship
Foster Homes and Foster Family Homes: Implications for Kinship
Foster Care as Family Preservation, Children and Youth Services
Review, Vol. 16, Nos. 1- 2 (1994), pp. 33- 63.

d U. S. General Accounting Office, Foster Care: Services to
Prevent Out- of- Home Placements Are Limited by Funding Barriers,
GAO/HRD-93-76 (Washington, D. C.: June 29, 1993). e F. H. Wulczyn
and R. M. Goerge, Foster Care in New York and Illinois: The
Challenge of Rapid Change, Social Service Review, June 1992, pp.
278- 94. f Mark F. Testa, Home of Relative (HMR) Program in
Illinois Interim Report (Chicago, Ill.: University of Chicago,
School of Social Services Administration, 1993). g For example, in
Georgia a child in kinship care is almost three times as likely as
a child in other foster care settings to remain in care for 1 year
or longer. U. S. General Accounting Office, Foster Care:
Children's Experiences Linked to Various Factors; Better Data
Needed, GAO/HRD-91-64 (Washington, D. C.: Sept. 11, 1991).

h Mary I. Benedict and R. B. White, Factors Associated with Foster
Care Length of Stay, Child Welfare, Vol. 70, No. 1 (1991), pp. 45-
58. i For example, children who entered kinship care in fiscal
years 1979 to 1980 were 25- percent less likely to be discharged
than children who entered other foster care settings before fiscal
year 1977, and children who entered other foster care settings in
fiscal years 1979 to 1980 were 14- percent more likely to be
discharged than children who entered other foster care settings
before fiscal year 1977. Discharge includes return to parental
custody, placement in private guardianship, adoption, or staying
in the child welfare system until age 18. Mark F. Testa, Kinship
Care in Illinois, in J. D. Berrick, R. P. Barth, and N. Gilbert
(eds.), Child Welfare Research Review, Vol. 2 (New York: Columbia
University Press, 1997), pp. 101- 29.

GAO/HEHS-99-32 Kinship Foster Care Page 64

*** End of document. ***