Date of Graduation
Eberly College of Arts and Sciences
Tracy L. Morris
Kevin T. Larkin
Amy E. Root
This study examined an observational method for identifying specific behaviors to indicate social anxiety and social withdrawal in preschool children. During an interaction task with a novel adult, 28 children (ages 4 and 5) were observed for specific verbal and non-verbal behaviors (i.e., total utterances, commands, questions, unsolicited conversation, direct responses, eye gaze aversion, non-verbal response, physical distancing, freezing, smiling ) and global behaviors (i.e., shyness, volume or speech, and postural rigidity), theoretically relevant to early childhood social anxiety. Behaviors observed during the interaction task were examined in relation to the parent-reported Spence Preschool Anxiety Scale's social anxiety subscale (Spence, Rapee, Edwards, & Ingram, 2001) and the percentage of solitary behavior observed during unstructured play at preschool (Morris, Messer, & Gross, 1995). Four years later, participants were re-contacted and 15 parents and children completed the Social Phobia and Anxiety Inventory for Children, Parent Report (P-SPAIC; Beidel, Turner, Hamlin, & Morris, 2000), the Child Behavior Checklist Social Competence Subscale (CBCL, Social Competence; Achenbach, 1991), and the Social Phobia and Anxiety Inventory for Children, Item 10 self-report (SPAIC-10, T. Morris, personal communication). Preliminary analyses revealed that the specific verbal and non-verbal behaviors (with the exception of physical distancing and commands) were significantly correlated with the global behaviors. The global behaviors did not predict a significant amount of variance in solitary behavior during unstructured play. None of the specific verbal or non-verbal behaviors were significantly related to solitary behavior during unstructured play or the Spence Preschool Anxiety Scale's social anxiety subscale. Although solitary behavior during unstructured play and the Spence Preschool Anxiety Scale's social anxiety subscale were significantly related, neither one was significantly differentially related to any of the specific verbal or non-verbal behaviors observed during the interaction task. Unsolicited conversation and direct responses were significantly related to CBCL Social Competence at follow-up. Although a significant difference was not detected, a large effect was seen for total utterances, commands, and non-verbal responses to direct questions for those children who were elevated on one or more of the follow-up measures compared to children who were not elevated on any follow-up measure. Although limited by a lack of adequate statistical power and a restricted range of social anxiety at baseline, the study has important implications for the identification of specific behaviors in preschool that may be risk-factors for social anxiety childhood.
Ale, Chelsea M., "Identification of Early Behavioral Markers of Anxiety and Social Withdrawal in Preschool Children" (2012). Graduate Theses, Dissertations, and Problem Reports. 447.