A dual pathway for understanding the relation between wellbeing and resilience

Resilience, the capacity to navigate adversity and adapt, operates at individual, family, and community levels. Indonesia, with its rapid urbanization and slum settlements prone to flooding (especially in Bima, Manado, and Pontianak), provides a crucial context for studying resilience. This study focuses on the resilience of Indonesian slum dwellers, particularly examining the role of wellbeing. Wellbeing is conceptualized using White's relational wellbeing framework, encompassing material, relational, and subjective dimensions, acknowledging the interconnectedness of these dimensions. Benard's perspective on individual resilience, emphasizing the importance of protective factors like family and community support, is also integrated. The study addresses the gap in research on the factors contributing to individual resilience among Indonesian slum dwellers, particularly from a relational wellbeing perspective. The main research question explores the extent to which family and community protective factors explain the relationship between relational wellbeing dimensions and individual resilience.

Existing literature highlights resilience's diverse applications and its multi-level nature. Individual resilience is linked to positive self-evaluation and coping mechanisms, particularly crucial in challenging environments. While subjective wellbeing (SWB) and psychological wellbeing (PWB) theories are relevant, the study adopts White's relational wellbeing framework, emphasizing the centrality of social relationships in shaping wellbeing. Benard's work emphasizes the role of protective factors (family and community) in fostering individual resilience. While previous studies have explored these factors, research on Indonesian slum dwellers, especially through a relational wellbeing lens, is limited. This study addresses this gap.

The study utilizes data from the RISE project, focusing on three Indonesian cities (Bima, Manado, and Pontianak) characterized by high flood risks and slum settlements. Due to incomplete population registries, a random walk method was used to sample 700 adults (aged 18 and above) residing in the selected areas for at least five years. Confirmatory factor analysis (CFA) using the lavaan package in R validated the measures for individual resilience (based on Benard's framework, ultimately reduced to two factors: social competence/problem-solving and autonomy/sense of purpose), relational wellbeing (six dimensions: family & community participation, security & competition, subjective material wellbeing, public facility satisfaction, religiosity, and self-concept), and protective factors (family and community, each with two dimensions: caring/expectations and opportunities/participation). A parallel mediation analysis was conducted to test the hypotheses, controlling for age, gender, and community resilience dimensions (assessed using a CDRI framework across physical, social, economic, institutional, and natural aspects). Individual characteristics like age, gender, education level, and income were also considered.

Significant differences in individual resilience levels were observed across the three cities, with Pontianak showing the highest level and Manado the lowest. Partial support was found for the hypothesis that relational wellbeing dimensions are positively associated with individual resilience, with family and community participation and public facility satisfaction showing significant positive correlations. Family protective factors positively mediated most relationships between relational wellbeing dimensions and individual resilience, except for religiosity. Community protective factors, however, demonstrated an inverted mediation effect, negatively mediating all relationships except that with religiosity. The effect size for most significant relationships was small, although some showed a medium effect size. A negative interaction was found between religiosity and the community protective factor on individual resilience, suggesting that high religiosity might decrease the positive effect of community involvement on resilience. Age showed a small negative correlation with individual resilience, while gender showed no significant relationship. Economic and natural dimensions of community resilience were positively associated with individual resilience.

The positive mediation effect of family protective factors suggests that positive assessments of family and community engagement, security, financial circumstances, and self-concept foster positive family interactions and enhance resilience. The negative mediation effect of community protective factors is unexpected and suggests that community involvement may not always positively influence resilience, possibly due to several factors: the multi-faceted nature of community support, the potential confounding influence of collective efficacy, and the possibility of non-empowering or dependency-inducing support. The negative interaction between religiosity and community protective factors highlights the potential for religious coping strategies to lessen reliance on community involvement in dealing with adversity.

This study advances understanding of the complex interplay between relational wellbeing, individual resilience, and the mediating roles of family and community. The findings highlight the importance of family support in fostering resilience and suggest the need for caution in assuming a universally positive relationship between community involvement and resilience. Future research should explore these relationships in rural areas, investigate specific community activities, investigate additional mediators, and incorporate longitudinal data to capture dynamic changes across time, especially during periods of flooding.

The study's limitations include its focus on urban areas, lack of detailed information on specific community activities, relatively small effect sizes for some significant relationships, and the cross-sectional nature of the data, which prevents drawing conclusions about changes during normal versus flooding periods. Future research should address these limitations.