Impact Findings from the Federal Multi-Site Evaluation of Foster Youth Services: Early Start to Emancipation Preparation-Tutoring Program

Background and Purpose: Education is widely recognized as an important contributor to success during the transition to adulthood, but research shows that most foster youth leave care with educational deficits and few obtain post-secondary education (Courtney et al 2007; Pecora et al 2005). Tutoring has been used to address educational deficits of adolescents and tutoring programs targeting foster youth have been developed recently. One of the programs included in the national evaluation of interventions funded through the Chafee Foster Care Independence Program is the Early Start to Emancipation Preparation (ESTEP)-Tutoring Program (LST) provided in Los Angeles County. ESTEP-Tutoring is provided primarily in the home of the youth being served. It is based on an individual learning model in which tutors are trained to assess student's skills in math, reading, and spelling and work with students at their own pace. The tutoring relationship is also intended to foster a mentoring relationship. Each youth is eligible for up to 65 hours of tutoring. This study examines the services provided by and impacts on selected outcomes of the ESTEP-Tutoring program. Methods: The study involved three in-person interviews over two years. The sampling frame was all foster youth referred for ESTEP-Tutoring during the study. Youth assessed by the program as being 1-3 years behind in either math or reading were randomly assigned to either the treatment or control group. The study sample consists of 445 youth referred to the program. The baseline response rate exceeded 95%, with over 90% of the baseline sample interviewed at the end of two years. 61.8% of all treatment youth started tutoring and on average, youth received 18 hours of math tutoring and 17 hours of reading/language tutoring. Outcomes assessed included Woodcock Johnson Tests of Achievement for letter-word identification, calculation, and passage comprehension. Youth were also asked about what grades they had received during their last semester in school, highest grade completed, and school behavior. Intent-to-Treat analyses were conducted to compare outcomes between the treatment and control groups. Additional impact analyses included regression models with baseline covariates and instrumental variable regression models that account for potential bias introduced by the presence of control group crossovers (12.3% of the control group received some ESTEP-Tutoring). Covariates in the analysis included physical and mental health, substance abuse, social support, and deviant behavior. Results: Educational outcomes were generally poor and across all outcomes there were no statistically significant differences between ESTEP-Tutoring and control group youth. Only 15% of the tutoring arrangements developed into long-term mentoring relationships. About three-fifths of the youths in the control group reported receiving educational tutoring from some source, more than twice as often from school as from anywhere else. Conclusions and Implications: The ESTEP evaluation raises questions about the appropriateness of home-based tutoring for foster youth, given the difficulty of tailoring such tutoring to skills being taught at school and youths' learning disabilities. In addition, the large percentage of foster youth who received tutoring from some source calls for more research on the impact of tutoring on foster youth.