Structural Equation Modeling of Separation Anxiety in Children Based on Maternal Anxiety, Mother-Child Relation, Maternal Attachment Style, Children’s Incompatibility Schemas and Child Attachment Style

Abstract

Child separation anxiety is one of most common anxiety disorder that strongly influence educational activities, social compatibility and adult anxiety disorders. However, all combination factors never studied in a fully integrated approach. This study aim to investigate some determining factors of such disorder, entirely. It was hypothesized that childhood anxiety can be predicted by a sequence of factors directly or indirectly. A sample of 620 mother-child pairs (6 and 7 years old girls and boys) were selected. Cattle Anxiety Scale (CAS) for Adults, Mother-Child Relationship Scale, Adult Attachment Scale, Child Attachment Assessment Scale, Children Incompatible Schema Questionnaire, and Child Symptom Inventory, using self-reported data were collected. Data was analyzed by Structural Equation Modeling (SEM). Relative fitness index for fitted model was 0.741 and incremental fitness index was higher than 0.9. In the standard model, a significant causal relationship between maternal anxiety and mother-child relation was observed. Mother-child relation, itself, in two paths, have effect on separation anxiety disorder (SAD) through intermediate variables of maternal attachment style, child attachment style and children incompatible schemas. One of present study limitations is the use of correlation plan and a causal interpretation of its results. The second limitations was the self-reporting feature of research tools that showed the results of perception of the participant from the measured structures. The measured structures may be influenced by different cultural, individual and family values. The findings of this study showed that except for the negative impact of two intermediate variables, namely mother-child relationship on maternal attachment style, in two directions of model, the positive coefficient relations indicate a positive effect on intermediate variables and finally dependent variable. DOI: 10.5901/mjss.2016.v7n4S1p54