Acknowledgements

1
Acknowledgements

• This study is a secondary analysis of an original
  study conducted by the Center for Families at the
  University of Maryland School of Social Work in
  collaboration with the Baltimore City Department
  of Social Services.

2
Purpose of Dissertation

• To identify personal and interpersonal factors
  that contribute to resilience of young adults who
  left out-of- home care
• To explore relationship between resilience and
  perceived levels of stress, social support,
  spiritual support, community support, and
  differences based on gender

3
Background

• Young adults in transition are a vulnerable
  sub-population of child welfare system
• Challenges associated with maltreatment (Bagley
  Mallick, 2000 Cohen, Brown, Smailes, 2001
  Fergusson Lynskey, 1997 Silverman, Reinherz,
  Giaconia, 1996)
• Challenges associated with out-of-home care
  (Courtney Barth, 1996 Jonson Reid, 2000)
• Overrepresentation of children of color (Sedlak
  Broadhurst, 1996 Roberts, 2002)
• Challenges associated with poverty (Brooks-Gunn
  Duncan, 1997 Duncan Brooks-Gunn, 2000)

4
Significance of Studying Resilience

• Young adults are a vulnerable group
  identification of factors related to overcoming
  adversity is important
• Resilience viewed as developmental process
  (Cicchetti Garmezy, 1993 Luthar, Cicchetti,
  Becker, 2000) period of transition is critical
  developmental stage
• Resilience research indicates positive
  relationship between social support and
  resilience (Masten Coatsworth, 1998 Werner
  Smith, 1992), but this relationship understudied
  among transitioning adults

5
Research Questions

• 1. What functions of social support do young
  adults perceive as available to them?
• 2. Does the young adults perceived availability
  of support from family members versus friends
  differ?
• 3. To what degree do young adults demonstrate
  resilience?
• 4. Do personal factors and interpersonal factors
  relate to degree of resilience?

6
Theoretical Framework

• Ecological Model
• Human development is life course representing
  individual pathways (Germain Gitterman, 1996,
  p. 21)
• Interchange of supports exists between individual
  and environment
• Resilience Theory
• Resilience dynamic process encompassing
  positive adaptation within the context of
  significant adversity (Luthar et al., 2000)
• Protective mechanisms assist youth to overcome
  adversity
• Cultural Competence Perspective
• Important to understand social/political/cultural
  context of challenges and strengths of youth in
  transition
• Facilitates interpretation of findings and
  implications for research, policy, and practice

7
Conceptual Model of Resilience

• Positive Outcomes
• High school completion
• Advanced education
• Employment

Protective Factors Individual (spiritual
support) Friends Family (social)
(social support)
(support) Community (social embeddedness, sense
of community)
Risk Factors Individual (stress) Friends
Family
(child (maltreatment) Community
(violence, poverty)
YOUNG ADULT (age, gender)

• Negative Outcomes
• Criminal activity
• Substance abuse
• Homelessness
• Early parenthood

(Murray, 2003)
8
Methodology

• Secondary data analysis of an original study
  conducted by Center for Families at University of
  Maryland School of Social Work in collaboration
  with BCDSS (DePanfilis Daining, 2003)
• Original study used a single cross-sectional
  research design to collect data about a sample of
  young adults who aged out of care

9
Study Site and Sampling

• Study Site - Baltimore City, Maryland
• Large urban area in Mid-Atlantic region of U.S.
• Population of 651,210 (U.S. Census Bureau, 2000)
• Approximately 11 between ages of 18 to 24
• 64 of Baltimore City is African American
• Sample Universe (N189)
• All youth (age 18 to 24) who left foster care or
  kinship care between October 1, 1999 and
  September 30, 2000

10
Procedures

• Recruitment
• Invitation letter mailed to eligible participants
• Interviews offered at 3 locations
• Home
• BCDSS local office
• Center for Families research office
• Data Collection
• Self-administered computer-assisted interview
• Five trained research staff facilitated
  interviews
• Followed standardized procedures outlined in
  manual
• Supervised by project manager

11
Eligibility of Interview Participants

• Total N 189
• Ineligible 18 (9.5)
• 3 deceased
• 14 incarcerated
• 1 institutionalized
• 171 individuals were eligible for the interview

1.6
7.4
.05
12
Participation Rates N 171

• Interviews between April 2002 and February 2003
• 103 Participated
• 15 Non-Responsive
• 7 Refused
• 46 Not Located
• 82.4 were interviewed of those located

60.2
8.8
4.1
26.9
13
Interview Sample Demographics

• 65 Female
• 90 African American
• 87 Never Married

14
Measure of Dependent Variable

• Resilience measured by composite score reflecting
  seven outcome domains
• work status
• high school (or equivalent) participation
• advanced education participation
• avoidance of early parenthood
• housing stability
• avoidance of drug use
• avoidance of criminal activity

15
Measures of Independent Variables
16
Data Analysis Plan

• Multiple Regression Analysis
• Dependent Variable Degree of resilience
• Covariates Age and Gender
• Independent Variables
• Level of perceived global life stress
• Level of perceived social support from family
• Level of perceived social support from friends
• Level of perceived spiritual support
• Level of perceived social embeddedness
• Level of perceived sense of community
• Current status Proposal defended in December
  2003 and presently conducting data analyses

17
Study Limitations

• Cross-sectional design prohibits causal inference
• 60 response rate means differences between
  respondents and nonrespondents may bias results
• Study population from one city limits
  generalizability of study findings
• There has been limited research done that tests
  validity of measures used with African American
  population

18
Study Strengths

• Use of self-administered computer-assisted
  interview may decrease response bias
• Use of standardized measures with known
  reliability and validity increases likelihood
  that results will represent constructs under
  study