| Abstract | It is important to identify youth who have a substance use disorder (SUD) when they enter the juvenile justice (JJ) system using a screener such as the TCU Drug Screen 5 (TCU DS 5), so that necessary treatments can be provided to them. While the TCU DS 5 is a valid, evidence-based screener, the use of an item response theory (IRT) model may better differentiate among mild, moderate, and severe forms of SUD. The current study analyzes the feasibility and incremental value gained in using an IRT model to compute total TCU DS 5 scores compared to its current scoring methodology. The results reveal that while IRT may not be worthwhile as the standard method of scoring, there are benefits to using IRT to assess the validity and value of individual items in a screening instrument. |
