The purpose of this dissertation is to analyze longitudinal impacts of parent attachment in early adolescence on adolescent development and to examine mediating effects by longitudinal changes of self-esteem and peer attachment. This study used latent...
The purpose of this dissertation is to analyze longitudinal impacts of parent attachment in early adolescence on adolescent development and to examine mediating effects by longitudinal changes of self-esteem and peer attachment. This study used latent growth model to analyze 5-year longitudinal data collected from the fourth grade in elementary school to the second grade in middle school by the Korea Youth Panel Survey.
The findings of this study are as follows; firstly, latent growth model was used to analyze major variables. Linear change model, which proved to be adequate in both the model fit and the parsimony fit, was adopted as a final model. In linear change modeling, parent attachment, self-esteem, peer attachment and academic achievement as positive determinants were on the decline, whereas depression/anxiety and delinquency as negative determinants were on the rise. Also, the first year intercept and the slope variance of variables were found to be statistically significant. This meant that the first year intercept and the five-year slope significantly made individual differences.
Secondly, this study examined the effects of early adolescent parent attachment on adolescent development. Parent attachment was proved to have longitudinal and cross-sectional influences. Specifically, the higher the intercept of parent attachment was, the higher the intercept of adolescent academic achievement was. However, the intercept of depression/anxiety and delinquency was lower. Also, the first year intercept of parent attachment had an significant impact on the slope of academic achievement and delinquency, except depression/anxiety. This meant that the higher the intial rate of parent attachment was, the more slowly academic achievement rate decreased and the delinquency rate increased. The analysis of relations between slopes indicated that the faster the rate of parent attachment decreased, the faster the rate of academic achievement decreased and the faster the rate of depression/anxiety increased.
Thirdly, this study scrutinized longitudinal mediating effects of self- esteem and peer attachment in the relations of parent attachment and early adolescent development. The mediating effects of self esteem and peer attachment were found to have a significant influence. As we expect, these mediating effects had a significant impact on increasing the intercept of positive determinants and reducing the intercept of negative determinants between parent attachment and early adolescent development. Also, longitudinally the decrease rate of positive determinants were slower and the increase rate of negative determinants were slower. However, the findings of this study with regards to the intercept and the slope of peer attachment in the relation of parent attachment and delinquency were inconsistent with the precedent studies and the assumptions of this study. This requires supplementary analysis and careful interpretations.
Fourthly, this study conducted multiple group analysis regarding gender, parent educational background and poverty. Depending on gender and parent educational background, the differences between the model with restrictions and the model with no restrictions were statistically significant. This means that this study model should be applied differentially according to gender and educational background.
Finally, this study discussed theoretical and practical implications with regards to attachment theories and adolescent development and suggested limitations of this study, which should be supplemented in future studies.