Dr. Tom George developed the WARNS and conducted initial validation studies at the Washington State Center for Court Research, an effort funded by a grant to the Center from the John D. and Catherine T. MacArthur Foundation. Development of the WARNS instrument began in the summer of 2008 in Washington State with discussions among juvenile justice professionals about the need for a risk and needs assessment for status offenders and low-level juvenile criminal offenders. The WARNS was modeled after the widely used and validated Washington State Juvenile Court Assessment for juvenile offenders (2004 Washington State Juvenile Court Assessment Manual by Robert Barnoski). However, the need for an easily administered instrument, with the potential for use with a large number of youth, led to the development of a brief self-report instrument which could be administered efficiently and economically on a large-scale basis.
The domain and item development process began with a review of the research literature on correlates and predictors of truancy, delinquency, and dropping out of school. This included a review of validated research and clinical instruments developed to assess juvenile offending, child and adolescent psychopathology, and educational engagement and outcomes. Based on these reviews, areas of assessment were identified that would appropriately address youths’ risks for problematic outcomes as well as their social and emotional needs. Finally, items were developed to assess these risk factors.
A series of studies were conducted in 2014 in order to build upon the initial pilot studies and to provide additional reliability and validity evidence to continue the support for the WARNS. Conducted by researchers at the Learning and Performance Research Center at Washington State University, these new studies provided strong evidence for the WARNS in identifying the needs of youth who are truant or at-risk for truancy.
Evidence demonstrated that the WARNS scores have:
- High internal consistency.
- A high degree of test-retest reliability.
- A high degree of inter-rater reliability.
Evidence also demonstrated that:
- Test content in the WARNS is supported.
- The WARNS factor structure is supported.
- The WARNS items function similarly across major identified groups.
- Correlations with external variables reveal that WARNS scores are correlated as expected with other measures.
- The WARNS scores show expected differences in expected groups.
Major forms of reliability and validity evidence were provided to support the WARNS scores and inferences. However, exhaustive validity evidence is impossible to collect given (a) building validity evidence is an ongoing process, (b) an assessment program or assessment developer cannot foresee all uses of the scores, and (c) a test developer cannot predict every type of child that may be assessed. However, the evidence we do present provides a strong foundation for the use of the WARNS.
Please see the WSU WARNS Technical Manual for detailed information on the validity and reliability of the instrument.