Date of Award
Master of Arts in Community Psychology, MA
Reed Mueller, Ph.D.
An ordinal regression method was used to model the relationship between the ordinal outcome variable, an ADHD index created for the purpose of this research, and the predictor variables of Race, Economics, Neighborhood Rating, and School Category/Type. A secondary data analysis of the Panel Study of Income Dynamics (PSID) and Child Development Supplement (CDS) was used in this study. The final sample consisted of 1,169 elementary, middle, and high school children ages 9-18. Six analyses were conducted to determine the best fit model for the data: a combined model with all the predictor variables, one for each individual predictor variable, and one model with the combination of race and neighborhood because they were the closest in showing value. The individual model using the variable on race was the only model that showed fit at a statistically significant level. Although the model was a good fit, the R2 was statistically small, which showed that race is likely to be a poor predictor of the outcome of ADHD severity for any student.