Theses and Dissertations

Date of Award

2020

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

College of Psychology

First Advisor

Ralph E. Cash

Second Advisor

Diana Formoso

Third Advisor

Ryan Black

Keywords

Eyberg Child Behavior Inventory (ECBI), factor analysis, Hispanic, Rasch modeling

Abstract

This dissertation was designed to confirm the factor structure and to assess the psychometric functioning of the Eyberg Child Behavior Inventory (ECBI) in an ethnically diverse clinical sample using Confirmatory Factor Analysis (CFA) and Rasch modeling. The sample included 221 children and adolescents (72% male and 28% female) whose mothers completed the ECBI. Related to ethnicity, 43.4% of the sample was Hispanic American (HA), 41.2% was European American (EA), 12.2% was African American, and 3.2% identified as “other.”

Dimensionality of the ECBI was explored using CFAs and by evaluating model fit criteria. An Andrich Rating Scale Model was employed to assess the rating scale functioning of the ECBI scales. The degree of item invariance across HA and non-HA groups was explored using differential item functioning. Reliability of the scales was assessed using Cronbach’s alpha, as well as Rasch-based estimates of reliability.

The results confirmed the superiority of the 3-factor model for the ECBI in an ethnically diverse sample. The 3 scales were found to be unidimensional measures of specific domains of child behavior and their items did not exhibit statistically significant invariance between HA and EA groups. Furthermore, the scales demonstrated acceptable reliability and good convergent and discriminant validity. The findings provided novel empirical support for the cross-cultural use of the ECBI scales and the generalizability of the findings related to the factor structure of the scales to populations with a large HA representation. Lastly, the results revealed that, for the ECBI scales, a 5-category rating scale is optimal for measurement.

Included in

Psychology Commons

Share

COinS