Objective Measurement of Sex Differences and Developmental Delays and Disabilities in Preschool Classroom Social Networks

• In Event: Using Sensing Technologies to Understand Children's Experiences in Early Childhood Education Classrooms

Abstract

Developmental delays and disabilities (DDs) are often characterized by deficits in communication and interaction and are more prevalent in boys than girls. Most research on DDs therefore focuses on boys, leaving questions remaining about whether girls and boys with DDs differ, especially in real-world contexts, like preschool classrooms. Here, we use automated sensing technologies to create preschool social networks and objectively measure how girls and boys with DDs differ from each other and their typically developing (TD) peers, both generally and compared to same sex peers, specifically in their language and social interactions.

Participants included 98 children (42 girls, 56 boys; 66 DDs, 32 TD) enrolled in 12 inclusive classrooms. During monthly observations, children wore lightweight vests equipped with Language ENvironment Analysis (LENA) recorders to measure vocalizations and two Ubisense tags to measure location/orientation. Ubisense measures of proximity/orientation were used to determine social contact and synchronized to LENA data to assess each child’s social language output to and input from peers. Each child’s language abilities were assessed using the Preschool Language Scales (PLS-5). Data were aggregated to create classroom social networks based on the frequency of dyadic language-mediated interactions (Figure 1). Social network analysis yielded measures of the modularity, or cohesiveness of groups (i.e., DDs/TD) within the class, and each child’s degree centrality to their network, or the social ties children have to their peers based on the sum of language input and output to and from peers (co-talk). We also assessed how children’s assessed language abilities influenced their interactions and role in the classroom network.

We compared group modularity, individual degree centrality, and the effect of assessed language abilities between girls and boys with and without DDs. Preliminary findings from 77 participants showed that in general, children with DDs had lower modularity than their TD peers (p = .04), but specifically TD girls had significantly higher modularity than any other group (p = .005). This finding could indicate that it is more difficult for girls with DDs to interact with TD girls, because the TD girl group is so modular on its own. For degree centrality, TD children were more central to classroom networks than peers with DDs (p = .01), indicating that TD children produce more output and receive more input from peers than children with DDs. Further, when accounting for PLS-5 scores, an interaction exists between TD boys and language abilities (p = .01), suggesting that language abilities may be especially important for supporting co-talk for boys, while language abilities matter less for girls (Figure 2).

Overall, the results suggest that there are DD-related group differences in social communication and interactions, but those group differences vary with sex. Using objective measurement from sensing technologies provides a clearer understanding of children’s social interactions, indicating that both group and sex are crucial components of classroom network formation. Such findings are made possible with the use of objective measures and may lead to important insights of differences between girls and boys with DDs that are relevant for diagnosis and intervention.

Authors

• Regina Fasano, University of Miami
  Presenting Author

• Laura Vitale, University of Miami

• Daniel Messinger, University of Miami

• Lynn K Perry, University of Miami