Factorial Validity and Measurement Invariance of the Communication and Symbolic Behavior Scales Developmental Profile

Individuals with developmental language disorders experience an array of challenges that persist from childhood into adulthood, highlighting the need for advancements in early detection and intervention for this population. The identification of young children who are at risk for persistent language delays continues to be fraught with instability and a reduced reliability. There is a continued need for robust and valid measures of early language and communication that can adequately characterize trajectories of communication development in very young children. This study examined the latent factor structure and measurement invariance of the Communication and Symbolic Behavior Scale Developmental Profile Caregiver Questionnaire (CQ), a measure of early social communication and symbolic behavior, in an archival sample of 3498 children aged 8-25 months ascertained through primary care screenings. A stepwise approach using confirmatory and exploratory factor analysis was implemented to compare the dimensionality and invariance of the CQ across six age groups. Results of confirmatory factor analyses showed hypothesized measurement models of the CQ evidenced borderline to poor fit across age groups, preventing further examination of measurement invariance across age. Exploratory analyses revealed the presence of problematic items characterized by low variance in responses and low endorsement of the Not yet response category. Problematic items were addressed through exclusion or modification of the item’s response format. Two patterns of extracted factors emerged during exploratory factor analyses, a two-factor model for the 8-10, 14-16, 23-25 months age groups and a three-factor model for the 11-13, 17-19, and 20-22 months age groups. A consistent sounds and words use factor was extracted across the four oldest age groups regardless of the number of extracted factors. The present study highlights directions for future research regarding the use of the CQ, including aspects of the measure that can be adjusted to improve its measurement of early social communication abilities in infants and toddlers.