Poster #57 - Differences in Parent and Child Report of the SCARED: Implications for Investigations of Social Anxiety

• In Event: 1-005 - Poster Session 01
  In Poster Session: PS 01 Section - Developmental Psychopathology

Integrative Statement

The Screen for Anxiety-Related Emotional Disorders (SCARED; Birmaher et al., 1997) is an instrument widely used to assess anxiety symptoms in childhood and adolescence. The SCARED provides summed scores for total anxiety, as well as five subscales (general anxiety, social anxiety, panic disorder, school avoidance, and separation anxiety). The SCARED is generally administered to both the child and their parent and has shown acceptable convergent validity with other measures of anxiety. However, child and parent reports are frequently only moderately correlated (Birmaher et al., 1997; 1999) with the social anxiety scale having the lowest concordance among the subscales.
Here, two multi-trait, multi-method (MTMM) structural equation models were employed to empirically determine if child and parent report on the SCARED differ at either age 10 (n=367) or age 12 (n=360). The MTMM model is a bi-factor confirmatory factor analysis that utilizes multiple reporter methods to model latent traits, creating more robust estimates. Indeed, the MTMM model showed that child and parent reports are significantly different from a perfect correlation at both age 10 and age 12. Moreover, we found that parent and child reports are more concordant at age 12 than age 10. Next, given the paucity of psychometric data for the commonly used social anxiety scale, we aimed to examine convergent validity of the social anxiety scale. We compared relations between extracted factors scores of social anxiety from the MTMM model and observed social anxiety with relations between raw child and parent report and observed social anxiety. We utilized observed measures of social anxiety from a speech task (structured social interaction with an unfamiliar peer) and a “Get to Know You” task (GTKY; unstructured social interaction with an unfamiliar peer). At both age 10 and age 12, parent report and factor scores functioned similarly when predicting the observed anxiety during the structured speech task, while child report was not predictive of observed anxiety during the speech. At age 10, neither child nor parent report differed from the factor scores in predicting observed anxiety during the GTKY unstructured task. However, at age 12, child report and factor scores functioned similarly when associated with observed anxiety during the GTKY unstructured task and were more predictive than parent report. Critically, at both age 10 and age 12, averaging child and parent scores of social anxiety performed just as well as the factor scores when correlating with observed anxiety during either the structured or unstructured social interaction. In sum, our findings demonstrate that parent and child reports of anxiety significantly differ, especially at younger ages. Moreover, parent and child reports of social anxiety differentially relate to outcomes of interest. Finally, our study suggests that both sources of information play a valuable role in assessing social anxiety within different contexts and are most predictive when combined.

Authors

• Maureen E Bowers, University of Maryland, College Park
  Presenting Author

• Lori Reider, University of Maryland, College Park
  Non-Presenting Author

• Santiago Morales, University of Maryland
  Non-Presenting Author

• George A Buzzell, University of Maryland, College Park
  Non-Presenting Author

• Natalie Miller, University of Maryland, College Park
  Non-Presenting Author

• Sonya V Troller-Renfree, Teachers College, Columbia University
  Non-Presenting Author

• Daniel Pine, NIMH
  Non-Presenting Author

• Heather Henderson, Department of Psychology, University of Waterloo
  Non-Presenting Author

• Nathan Fox, University of Maryland
  Non-Presenting Author