Purpose: Federal Adoption and Safe Families Act has led to increased attention to the length of stay of children in foster care. The Child and Family Services Review process calls upon states to reduce lengths of stay or face fiscal sanctions. Because reunification is an important determinant of length of stay, the rates and timing of reunification are much studied in the child welfare literature. Factors such as age, reason for entry and ethnicity have been found to affect reunification. Minority children have been disproportionately represented in foster care, both in rates of entry and in length of stay. However, the identification of factors associated with reunification remains incomplete. Language is one such factor that has yet to be fully examined.

Methods: This retrospective longitudinal study examined the effect of maternal language on reunification rates and time to reunification of a cohort consisting of 15,144 mothers of California children who entered child welfare supervised foster care in 2001 using data from the Child Welfare Services Case Management System (the California SACWIS). To assure independence, when the mother had more than one child entering care in the year, reunification and other child data was based on the youngest child having complete information.

Results: Multivariate and time series analysis provided evidence of significant differences among the timing and rate of reunification of English speakers compared with non-English speakers. Having a primary language other than English was significantly associated with higher odds of reunifying and with shorter times to compared to English, after controlling for other factors including ethnicity, removal reason, age of child at removal, mother's age at removal, type of foster care, and drug and alcohol use.

Implications: These findings suggest that language is an additional factor that must be considered in understanding reunification. Language may be a proxy for acculturation. Implications for research include the need for further understanding of the sociocultural factors affecting child welfare and other social service. Finally, these findings demonstrate the utility of using administrative data for research purposes.