Reliability of Parent and Teacher Perspectives on Child Functioning in a Large Urban Sample of Children with ASD

Background:  

Cross-informant ratings are considered best clinical practice when assessing child functioning. However, data overwhelmingly support significant variability in ratings across informants. While the majority of this research has focused on typically developing children, recent work examining cross-rater agreement in ratings for children with ASD has found similar variability in ratings of functioning (Stratis & Lecavalier, 2014). The majority of this work has focused on emotional and behavioral problems; little work to date has examined parent/teacher ratings of children’s functioning more broadly, especially within a large-scale, diverse sample.  

Objectives:  

The purpose of this study is to systematically evaluate agreement between parent and teacher ratings of child functioning using standardized rating scale assessments. 

Methods:  

Participants in the current sample were part of a large study examining the effectiveness of a naturalistic behavioral intervention in school settings. Participants included 289 children, 3-11yrs old (M = 4.08, SD =3.21) who were served for ASD in public school programs. Participant measures were collected in the beginning and end of the school year. Parents and teachers of participating children completed the Vineland Adaptive Behavior Scales- II (VABS) and Pervasive Developmental Disorder Behavioral Inventory (PDDBI) at each time point. 

Results:  

Cohen’s kappa was used as a measurement of agreement between teachers and parents. Overall, data indicate only slight agreement between teacher and parent ratings across domains of each assessment, with Kappa’s ranging from .03 to .10. On the PDDBI, there was agreement on 33.3% of domains at both the beginning and end of the school year. On the VABS, agreement was higher at the beginning (43.7%) compared to the end (37.5%) of the school year. Parent and teacher ratings were significantly reliable (p < .04) on 18.8% of VABS domains and 20% of PDDBI domains at both time points.  When specifically considering these domains, three domains (PDDBI Expressive Language and Expressive Social Communication Abilities Composite and VABS Written) showed improved parent/teacher agreement from Fall to Spring. In contrast, three domains (PDDBI Semantic/Pragmatic Problems and VABS Play and Leisure skills and Fine Motor skills) showed declines in agreement from Fall to Spring. Potential moderating variables will be discussed.

Conclusions:  

The reliability in parent and teacher ratings of various domains of children’s functioning is highly variable. Data suggest only slight agreement between raters as well as an inconsistent pattern of change in agreement over time. Overall, these results support the larger literature demonstrating variability in cross-informant ratings of children’s behavior. The significant variability in teacher and parents’ ratings of functioning for children with ASD seen here warrants further exploration.