Measuring Play Constructs Across Measures in Young Children with ASD: Context Matters

Background:  Play is an important developmental skill that fosters appropriate language skills, provides opportunity for social interaction, and increases knowledge. Additionally, play is a vital component of developmental interventions in children with ASD (Ingersoll & Dvortcsak, 2009; Lifter, 2000). While play is included in diagnostic and developmental measures, no research, to date, has examined the relation of play constructs across measures. 

Objectives:  The aim of the current study was to examine play in toddlers and preschoolers with ASD.  Specific goals included: (1) to explore how play is assessed across five standard measures and whether play scores on these measures are related; and (2) to assess how play related to children’s developmental level. 

Methods:  Eighteen participants (M=34 months; SD=5.82; range: 26-55 months; 13 males; 5 females) with ASD participated in a diagnostic assessment.  The following measures were obtained: (1) ABAS-II: Adaptive Behavior Assessment System- Second Edition, (2) ADI: Autism Diagnostic Interview, (3) ADOS- Autism Diagnostic Observation Schedule, (4) CARS2-ST: Childhood Autism Rating Scale- Second Edition, (5) CSBS-ITC: Communication and Symbolic Behavior Scales Developmental Profile, and (6) MULLEN- Scales of Early Learning. For each measure, play constructs were developed, which were comprised of items that examined play.

Results:  Pearson’s correlations were conducted to examine whether play constructs were related to each other. Results revealed that play constructs for ABAS-II, CARS2-ST, and ADI were significantly related with CSBS-ITC, respectively (r=.75, r=-.54, r=-.66). Additionally, play constructs from ABAS-II and ADI were significantly correlated, r=-.57, p=021. There was a less clear relation between observational measures of play with no relation found between play constructs measured by the ADOS and the CARS (r=.11).   

Pearson’s correlations were conducted to examine the relation between play constructs and developmental level, as measured by scales from the MULLEN. Results revealed that fine motor skills were significantly correlated with the majority but not all play constructs (r’s ranging from .59 to .71) Visual reception was significantly correlated with the play construct from CSBS-ITC, r=.73, although visual reception was not related to other play measures. Receptive language skills were significantly related with play constructs from ABAS-II and CSBS-ITC (r=.51, r=.60). Lastly, expressive language was significantly correlated with play constructs from ADOS and CSBS-ITC (r=.74; r=-.59).

Conclusions:  Results revealed strong correlations between some play constructs. Although these data revealed that play measures were significantly related, the context differed across measures (e.g., observed play vs. caregiver reports), which can contribute to methodological differences in measurement of play. Further, while play constructs were significantly related to children’s developmental levels, this relation was also influenced by the context of the rating. Overall, it is important to measure play in a variety of contexts to determine the best possible play based intervention for children with ASD.