Representations of family relationships in children living with custodial grandparents

Abstract

Children's representations of family relationships were examined in relation to grandparent responsivity and depression, family sociodemographic risks, and children's behavior problems in custodial grandparent families. Using multiple methods, data were collected from 79 families with children aged 3 to 7.5 years. In 37 families, children were raised by grandparents as the result of maternal incarceration and, in 42 families, children lived with grandparents because of other parental problems. Results indicated that children with representations of less optimal family relationships and children whose grandparents were less responsive to them at home were more likely to exhibit externalizing behavior problems. In addition, children were more likely to include relationship violence in their story stem enactments when grandparents reported elevated depressive symptoms. Results suggest links among grandparent depression, quality of children's family relationships, and children's behavior problems in grandparent-headed households.

Keywords:

• attachment
• grandparent
• incarceration
• relationships
• children
• risk

Notes

1. When we conducted each analysis separately using ASCT mother and grandparent scores (rather than the total ASCT score), we found similar results.

2. Because of the association between gender and ASCT scores, we also conducted the regression analyses separately for boys and girls. Our results were similar to those presented above, although slightly attenuated because of the smaller subsample size. Two unique findings emerged. For girls, higher verbal reasoning scores predicted more optimal ASCT process codes (i.e., better coherence, less avoidance, etc), whereas for boys, age and HOME responsivity scores predicted more optimal process codes. In addition, for boys, responsivity in the home was more strongly associated with externalizing problems, whereas for girls, ASCT scores showed a stronger relation to externalizing problems, similar to the results of von Klitzing, Kelsay, Emde, Robinson, and Schmitz (2000). These analyses are available from the authors upon request, although they should be seen as exploratory in nature because of the small subsample size.