Youth at risk of emotional and behavioral challenges demonstrate academic difficulties in secondary settings. Parental involvement in their child's education is one method which contributes to improved academic outcomes. However, limited research exists on the use of measurement models to operationalize parental involvement with secondary students at risk for emotional and behavioral issues. The goal of this study was to examine the measurement properties of an existing three-domain conceptualization of parental involvement including school-based involvement, home-based involvement, and academic socialization, and a six-domain conceptualization to (a) determine whether these models are replicable with a population of secondary students at risk for emotional and behavioral issues, (b) examine the measurement invariance of the confirmatory factor analysis (CFA) models between samples of ninth graders in public and private schools without disabilities and ninth graders at risk of emotional and behavioral issues, and (c) determine whether parental involvement differs due to at-risk status. The six-factor model fit the data closest for both groups and was invariant between groups, suggesting this framework is appropriate for a variety of student groups. There were significantly lower levels of parental involvement in four of the six domains for the ninth-grade students at risk in comparison with students in the general population without disabilities.