Self-compassion and depressive symptoms in adolescence: the role of parents, peers and memory biases
Item statusRestricted Access
Embargo end date30/11/2021
Klases, Antonia Anette Maria
It is now widely acknowledged that self-compassion has benefits on mental health. Despite the growing evidence base for the significance of this construct as well as the growing increase in popularity of self-compassion research in general, there remains a lack of understanding of the development of self-compassion within adolescent populations. Furthermore, there is an extensive interest in assessing and enhancing compassion, however, an appropriate measure to evaluate how individuals experience compassion for others and compassion for self as well as other compassion aspects (e.g. receiving compassion from others) is lacking for this younger age group. This thesis aimed to take an initial step toward addressing these issues. Three empirical studies, each using a quantitative cross-sectional design were conducted to explore adolescents’ experience of relational compassion and to better understand the development of self-compassion within the adolescent population. Three independent samples were recruited in the present research. The sample for Study 1 (N= 609) and the sample for Study 2 (N= 191) consisted of adolescents aged 12 to 18 years recruited from six secondary schools in Scotland. The sample for Study 3 consisted of parent-adolescent dyads (N= 20) which were recruited via two separate routes: recruitment from secondary schools and recruitment via online platforms. The first empirical study examined the psychometric properties of the Relational Compassion Scale (Hacker, 2008) for the measurement of relational aspects of compassion in adolescents. The validation process of the RCS included using an exploratory factor analysis to identify the factor structure of the scale, assessing the internal reliability as well as construct, convergent and predictive validity of the scale. The findings revealed that the 21-item relational compassion scale is a psychometrically sound and theoretically valid measure for adolescents. The second and third study investigated the origin of self-compassion in adolescents, exploring in particular the influence of parental, interpersonal and cognitive factors. Using correlation and regression analyses, the second study found that adolescent avoidant attachment to parents and attachment to peers were significant predictors of adolescents’ levels of self-compassion. The cognitive factor, adolescents’ self-referential memory, did not emerge as a significant predictor of self-compassion. A large effect size was revealed for the relationship between self-compassion and depressive symptoms. The third study was an exploratory study aiming to provide a more in depth investigation of the role of parental factors in adolescents’ development of self-compassion. Statistical correlation and mediation analyses indicate that the level of self-compassion in parents is related to the level of self-compassion in their adolescent children. Parental criticism and adolescents’ anxious attachment style did not function as mediating mechanisms of this relationship. Overall, the present research provides an interesting insight into the understanding of the origin of self-compassion in adolescents and specifically offers a small step towards identifying potential developmental factors. The findings of this thesis further highlight the significance of investigating the concept of self-compassion and relational compassion in adolescents and add to the limited research literature in this field. In future research, it is important to focus on what other factors (e.g. cognitive processes) contribute to the development of self-compassion in adolescents. Future studies may benefit from using longitudinal and qualitative methods. Furthermore, the development of an age appropriate compassion measure for adolescents is needed. Overall limitations of the empirical studies in this thesis are discussed in terms of sample size and usage of self-report measures. Recruitment challenges of research participants, in particular of parent-adolescent dyads, are also acknowledged.