Youth Integration in Foster Care Settings: A Growth Curve Analysis

Background:
Youth integration into foster home settings is an important, yet under-investigated dimension of child
welfare research. Understanding the conditions under which youth integrate into substitute care holds
the potential to inform assessment and intervention strategies, and may promote the well-being of
youth in care.

The objective of this study was to examine how a youth’s sense of foster home integration changes over
time, and determine if youth/case characteristics explain growth trajectories for this construct. The
research question guiding this investigation was: Do youth characteristics (i.e., age, race, gender) and
case characteristics (i.e., prior foster care placements, placement changes, treatment exposure) explain
growth trajectories for youth reports of foster home integration?

Methods:
Data were drawn 202 participants in the Supporting Siblings in Foster Care (SIBS-FC) study (McBeath et.
al., 2014), a sibling-focused intervention designed to enhance well-being for youth in care (Kothari, et.
al, 2014). Participants were randomly assigned to receive the treatment condition or services as usual.
Measurements were taken at four separate time points every six months, over an 18-month period of
time.

The dependent variable in this investigation was a 9-item youth-reported measure of foster home
integration (alpha=.86). This measure asks youth to rate the quality of their interactions with caregivers
and other household members (e.g., How often did you talk to your foster parent?; How often are you
included in family decision-making?).

Independent variables included youth age (mean=11.8, SD=2.0), gender (50% female, n=101), and youth
race (50% non-White, n=101), as well as number of placement experiences prior to study enrollment
(mean=4.1, SD=3.2), placement changes during study participation (mean=.7, SD=1.1, range=0-7), and
treatment condition.

The analytical approach involved testing a series of growth curve models in Stata 14. An unconditional
means model was tested, followed by a repeated measures unconditional growth model for the home
integration construct. To test for the possible influence of youth and case characteristics on home
integration trajectories, the aforementioned independent variables were entered into a conditional
growth curve model.

Results:
The intercept for the unconditional means model (Yti=8.68, p<.01) and unconditional growth model
(Yti=8.71, p<.01) were significant. The intercept for the conditional growth curve model was also
significant (Yti=9.62, p<.01, Wald X2=20.27), with youth age (Tti=-.10, p<.01) and placements experienced
during study enrollment (Tti=-.12, p<.01) explaining individual variance in home integration over four
time points. Gender, race, treatment exposure, and pre-study placement experiences were not
significantly associated with home integration over time.

Implications:
Findings from this investigation suggest that being older and experiencing more placement changes was
negatively associated with home integration scores over time. This suggests the need to provide
additional home integration supports to youth who experience placement change, and to conduct
differential assessment of youth integration based on the developmental stage of the child. Examining
home integration may provide intervention researchers, practitioners, and policymakers with important
information about the experiences of youth in substitute care and help inform future research with this
population.