Title: An Evaluation of the Truancy Assessment and Service Center (TASC) Early Truancy Intervention Program with a Regression Discontinuity Design: A Replication Study

Background and Purpose: The Truancy Assessment and Service Centers (TASC) intensive case management intervention for elementary-aged students was assessed for effectiveness through a Regression Discontinuity (RD) design in one urban site for the 2006-2007 school year, one year after Hurricane Katrina. Results of the study indicated that TASC was effective, however, less so for African American children and children who were previously retained. The purpose of this study was to replicate the RD design in the same urban TASC site for the 2009-2010 school year, to assess whether the findings of the first study remained the same or whether a change occurred for students not affected by a major disaster.

Methods: A RD design was replicated with a second cohort of students (N= 890). Based on a pre-determined cut-off score, approximately half of the sample (47%) was assigned to the intensive case management intervention, while the other half received a notification letter and warning (53%). Both groups were monitored for the remainder of the school year.

Results: The replicated RD design yielded similar results to the original study. In both studies approximately 90% of participants were non-White, half were male (54%), and more than half were in Kindergarten and first grade (58%). However, more students in the replication study had been retained (24% vs 18%) and slightly more were in special education (11% vs 9%). In the replication study TASC reduced truancy almost 3% compared to 2% in the original study. The same correlates from the original study were included in the replication: gender, previous grade retention, case outcome, non-White race, special education status, grade, and an interaction term. The RD replication model proved to be significant (p <.001) with slightly more of the variance in the change in the rate of truancy explained by the replication study than in the original study (25% vs 21%). In both the original and replication study, truancy rates among participants in the control group remained at the pre-intervention levels, while truancy rates among those in the treatment group significantly declined (6.6% vs 7%; p < .01). Cases in the truancy intervention group that were successfully closed showed a 10% reduction in truancy compared to those that were not (p < .001) in both studies. In both studies TASC successfully reduced truancy overall, but is disproportionately effective across participants with the most at-risk students receiving less of a benefit from the program (p <.05), with students scoring 2 SD above the cut-off score receiving no benefit from the intervention. Finally, the intervention proved to be less effective with non-White children (p <.01) and with children who were previously retained (p <.01) in both studies.

Conclusion: This study supports the use of intensive case management services for children at risk for continued truancy in a school social work setting. The next steps in the research are to include an independent variable that combines multiple cohorts to assess any cohort effects. More research is also needed to assess which parts of the case management model are the most effective.