Measuring School-Wide Needs while Simultaneously Measuring Student Level Outcomes

Purpose: According to a recent fact sheet published by The Center for Health and Health Care in Schools, of the population ages 9 – 17, an estimated 21 percent experience the signs and symptoms of a DSM-IV disorder during the course of a year, 11 percent experience significant impairment, and 5 percent experience extreme functional impairment. Unfortunately, on average, only one-fourth of children in need of mental health services get the help they need. The critical role of schools in detecting and helping to treat youth with mental health needs is clear. In fact, research suggests that schools function as the de facto mental health system for children and adolescents. Mental health issues represent real barriers to learning for many students so it is in schools best interest to be active partners in ensuring that students in need get help. This presentation will focus on the development of a set of freely available tools and instruments that schools and community partners can use to assess and estimate student mental health needs. Implications for use needs data for planning services also will be discussed.

Methods: For the past five years, The Ohio Department of Education and the Ohio State University College of Social Work have partnered to develop and pilot innovative approaches to help Ohio schools and community partners address mental health or other critical needs of students. A significant component of this work focused on the identification and development of measures that could be readily used in a student needs assessment. Development of these measures were guided by important work in the areas of mental health, risk and protective factors, resiliency, and positive youth development. A battery of scales emerged from this work that have been used in a variety of needs assessments in Ohio schools. Data from these applied efforts have been invaluable in examining the psychometric properties of scales and in their refinement.

Results: In this presentation, we will share the results of scale structure and psychometrics for each of our set of scales. Our cross-site work has allowed us to conduct both exploratory factor analyses and, more importantly, confirmatory factor analyses. For example, we constructed a 10 item academic self-efficacy scale for use in a middle school needs assessment. With a sample size of 344 students, we found that the scale had a one-factor solution (principal axis factoring and varimax rotation) with factor loadings ranging from .56 to .81. Alpha for the scale was .90. We then used that scale in a second middle school needs assessment and used a confirmatory factor analysis approach to see if the scale structure held up in this different setting. Using EQS, we found reasonable fit (N = 387, CFI = .98, RMSEA = .055 with a 90% confidence interval of .038 - .072). Similar results were obtained for other measures.

Implications: Schools and community partners need free, accessible, and psychometrically sound measures for assessing student needs and outcomes. This battery of instruments will help to address that measurement need.