Missing data are nearly always a problem in research, and missing values represent a serious threat to the validity of inferences drawn from findings. Increasingly, social science researchers are turning to multiple imputation to handle missing data. Multiple imputation, in which missing values are replaced by values repeatedly drawn from conditional probability distributions, is an appropriate method for handling missing data when values are not missing completely at random. However, use of this method requires developing an imputation model from the observed data. This is typically a rigorous and time-consuming process. To encourage wider adoption of multiple imputation in social work research, a simple framework for designing imputation models is presented. The framework and its ability to generate unbiased estimates are demonstrated in a simulation study.

The child welfare system is an access point for children’s mental health services. Psychiatric residential treatment facilities (PRTFs) are the most restrictive, and most expensive setting for children to receive long-term care. Given the high rates of behavioral health concerns among maltreated children in out-of-home care, research is needed to examine the factors that predict entry in PRTFs among children investigated for maltreatment. This exploratory study used cross-sector administrative records linked across multiple systems, including child welfare records and Medicaid claims, from a single state over a five-year period (n = 105,982). Cox proportional hazards modeling was used to predict entry into a PRTF. After controlling for many factors, PRTF entry was predicted by diagnosis code indicating a trauma-related condition, antipsychotic medication prescriptions, and entry into lower levels of out-of-home care, supporting the view that youth are admitted to PRTFs largely due to clinical need. However, PRTF admission is also associated with characteristics of their experiences with the social service system, primarily foster care placement stability and permanency. Implications for practice and research are discussed.