Paper
Family stress and behavioural problems in kinship and non-kinship foster care. Report of a Flemish study
- issue: Issue 1 / 2009
- authors: Frank Van Holen, Johan Vanderfaeillie, Charlotte Haarsma
- keywords: Belgium, foster care, family, stress, behaviour, problems
- views: 4955
- downloaded: 0
- download pdf (75 Kb)
abstract
Background. Foster-care placements can be classified on the basis of the relationship between the foster child and the foster family. We make a distinction between 'non-kinship foster care' which refers to formal foster parenting by non-relatives and 'kinship foster care' which refers to formal foster parenting by relatives (family members or acquaintances). The percentage of kinship fostering is on the increase. This is an international trend. There is a good deal of discussion regarding the advantages and disadvantages of the two types of foster care. Empirical research on the subject is, however, scarce.
Research question. Does the relationship between the child and the foster family prior to the placement influence the family stress experienced by the foster parents or is family stress rather the result of other factors?
Methods. In a quantitative, exploratory research study we enquire amongst Flemish kinship and non-kinship foster families with a foster child between 6 and 12 years old who has been with them for at least 6 months. We give 2 questionnaires to the foster parents:
- using the Nijmegen Questionnaire for the Parenting Situation (Wels & Robbroeckx, 1996) we measure the subjective family stress;
- using the CBCL/6-18 we measure the emotional and behavioural problems of the foster child. We make use of the Total, the Internalising and the Externalising Problem Scores.
In addition a number of demographic and other background details were collected from the files. In 70 of the 96 foster families approached, all the data obtained from the foster mother was usable for analysis (response rate of 73%). Using traditional data analysis, it is verified whether kinship and non-kinship foster mothers differ according to demographic features and the experienced family stress and behavioural problems. Using 2 logistic regression analyses, a study is made of the degree to which the type of placement (kinship vs. non-kinship), among other variables, makes a unique contribution to family stress.
Key findings. The foster mothers report serious behavioural and emotional problems amongst 1/3 of the foster children. Foster children from kinship and non-kinship families do not differ according to the behavioural and emotional problems as reported by the foster mother (Total Problem Score: U=459,5; p=0,25 / Externalising Problem Score: U=496; p=0,49 / Internalising Problem Score: U=414; p=0,09). Kinship foster mothers do not differ, as a group, from non-kinship mothers when it comes to the family stress they experience (U=523; p=0,72).
Two logistic regression analysis show that the type of placement does not make any specific contribution to the family stress experienced. Each logistic regression analysis comprises two steps. In the first is controlled for the gender and age of the foster child, the age of the foster mother, the total number of children living in the family home and the CBCL problem scores (the Total Problem Score in the first and the Internalising and Externalising Problem Scores in the second analysis). In the second step the type of placement is added to the model.
Step 1 of the first logistic regression analysis results in a significant model (Nagelkerke R²=0,30; c=17,53; df=5; p<0,005). Only the Total Problem Score makes a unique contribution to the family stress (Exp(β)=1,04; p<0,01). Adding the type of placement, the model remains significant (Nagelkerke R²=0,31; c=18,29; df=6; p<0,01), but this step does not result in a better model (Blok fit c=0,76; df=1; p= 0,39).
The model of the second logistic regression analysis is also significant in the first step (Nagelkerke R²=0,34; c=20,85; df=6; p<0,005). Only the Externalising Problem Score makes a unique contribution to the family stress (Exp(β)=1,12; p<0,01). The model remains significant after inclusion of the type of placement (Nagelkerke R²=0,36; c=21,80; df=7, p<0,005) but it does not improve (Blok fit c=0,95; df=1; p= 0,33).
We conclude that only the CBCL Total Problem Score and the Externalising Problem Score make a significant contribution to the family stress experienced. Family stress is therefore predominantly explained as a result of the problem behaviour of the foster children and, to a lesser degree, by the nature of the placement. Other variables such as the gender and age of the foster child, the age of the foster mother and suchlike do not explain family stress either. Furthermore, it appears that only externalised problem behaviour is associated with family stress. This seems logical. After all, externalised behaviour is, in the first place, disruptive for the environment whilst internalised behaviour causes stress mainly for the child.
Implications and recommendations. The discussion is to be shifted from the advantages and disadvantages of kinship vs. non-kinship foster care to the importance of good matching, the starting point for which is the needs of the foster child and the specific family forms which can best respond to the parenting assignment. The idea is supported that it is not the type of foster family but other family characteristics which must be used as criteria in matching children with foster families.
Our results highlight the importance of the early recognition of problem behaviour. Early identification can make it possible to take measures with the aim of reducing the likelihood of overburdening foster parents and of breakdown. Some researchers focus on the seriousness of problem behaviour and develop taxonomies. This may give indications for differentiated diagnostics and treatment and may help sharpen up assessment and improve matching between foster child and foster family.
Longitudinal research into combinations of certain types of problem behaviour with specific family characteristics/situations should yield more valid indications, resulting in better risk-assessment and matching in the provision of services.
Key references
Wels, P. M. A., & Robbroeckx, L. M. H. (1996). Nijmeegse Vragenlijst voor de Opvoedingssituatie. Handleiding. (Nijmegen Questionnaire for the Parenting Situation. Handbook). Lisse: Swets & Zeitlinger.
Contacts: Frank Van Holen, Foster Care Facility 'OPVANG', Jetsesteenweg 603, 1090 Brussels, Belgium, E-mail: frank.vanholen@opvang.be, Phone 0032 2 4288009.