Assessing The Ability of the SRSS-IE to Accurately Predict Early Warning System Data in Elementary School Students
Author ORCID Identifier
Date of Award
Master of Public Health (MPH)
Dr. Brian Barger
Dr. Emily Graybill
In recent years there has been an increase in the reported cases of adolescent behavioral and emotional problems (Gresham, Lane, Macmillan, & Bocian, 1999). There is also disproportionate number of students from historically marginalized communities dropping out. Research has shown that high school dropouts experience higher rates of unemployment, higher rates of incarceration and more negative health outcomes when compared to those that graduate high school (Camacho-Morles et al., 2021). Often, adults struggle with behavioral health issues that first arose in their youth. Children with behavioral health issues can go on to experience greater mental health struggles later in life which can negatively impact interpersonal relationships, academic achievement, and employment status (Bevilacqua et al., 2018). In order to prevent a greater diminished quality of life as a result of mental health issues and/or dropping out of school, it is imperative to identify children with behavioral health issues early on to administer the necessary therapy, medication and/or other resources (Oakes, Lane, & Ennis, 2016). Systematic screening is largely the method by which children are identified as being at risk for emotional and behavioral problems (K. Lane et al., 2018). For example, the Student Risk Screening Scale (SRSS-IE) is a screening tool used to assess elementary school students for their risk for challenging and antisocial behavior based on Internalizing and Externalizing subscales (K. L. Lane et al., 2012). To assess the relationship between student demographics (e.g., race and gender) and externalizing/internalizing symptoms on early warning measures (EWS), this reported analyses conducted an iterative (i.e., measurement model for factors and then non-latent predictors and outcome. Behavioral screening data, EWS data, and sociodemographic data from 6644 elementary school students (grades K-5) from a school district in a southeastern state were utilized in this analysis. The measurement models utilized a two-factor model for externalizing and internalizing behaviors measured by the SRSS-IE. These two latent variables of externalizing and internalizing behaviors were then fitted to a structural equation model with race and gender serving at exogenous predictor variables, and attendance record and ODRs were endogenous predictor variables. Behavior outcomes, tardies, and absences were recorded with count data. Due to the zero-inflated nature of the data when it comes to our primary outcomes of ODRs, absences, and tardies; outcomes were bucketed into different respective discrete categories. Results of this analysis indicate that externalizing factor models had more significant associations to EWS data when compared to internalizing factor models. Contrary to past literature there was little to no direct associations between sociodemographic variables (race, grade, and gender) and ODRs. Overall, the findings of this study suggest that a two-factor model could implemented within EWS systems to better identify students at-risk for dropping out and/or emotional-behavior problems. Further research should be conducted in populations that have greater potential for generalizability.
Moore, Quentin, "Assessing The Ability of the SRSS-IE to Accurately Predict Early Warning System Data in Elementary School Students." Thesis, Georgia State University, 2023.
File Upload Confirmation
Available for download on Thursday, May 01, 2025