China's demographic shift, characterized by declining birth rates and increasing life expectancy, coupled with societal changes and evolving concepts of filial piety, has significantly impacted intergenerational relationships. The traditional model of family care, where older adults co-reside with their children, is increasingly less common, resulting in a growing number of older adults living alone. This trend is particularly pronounced in China, where the 'one-child policy' and urbanization have further strained traditional family support structures. The implications are significant, as older adults living alone face increased risks of health problems, depression, and reduced access to healthcare. This study aims to address the gap in research focusing specifically on non-co-residential intergenerational relationships from the perspective of adult children in China. It leverages the theory of intergenerational solidarity, incorporating dimensions of affectual, associational, functional, normative, and structural solidarity to understand the diverse forms of support and their determinants.

Research on intergenerational relationships, while extensive in Western academia, has a comparatively shorter history in China. Early work highlighted the differences between Western and Chinese models. The theory of intergenerational solidarity, initially proposed by Bengtson et al. (1976), and subsequently refined, provides a framework for analyzing the multifaceted nature of these relationships, encompassing affectual, associational, consensual, functional, normative, and structural dimensions. Numerous studies have employed latent class analysis (LCA) or latent profile analysis (LPA) to identify distinct types of intergenerational support and relationships, revealing varying patterns across different cultures and contexts. However, existing research often focuses on the older adult perspective or lacks a detailed examination of non-co-residential relationships from the viewpoint of adult children in China, particularly concerning the unique characteristics of this population given China's cultural emphasis on kinship and geographical proximity.

This study uses data from the 2017 CGSS, a large-scale national survey in China. The sample was selected to include adult children who do not co-reside with their parents. After excluding cases with missing data, the final sample comprised 1270 adult children. Latent profile analysis (LPA) was conducted using Mplus 8.0, employing 24 indicators across five dimensions of intergenerational solidarity (excluding consensual solidarity due to data limitations and the focus on adult child perspectives). The LPA identified distinct profiles of intergenerational relationships. Subsequently, multinomial logistic regression models, implemented using IBM SPSS 21.0, examined the effects of demographic (gender, living area, education level, marital status) and personal (annual income, self-rated well-being, self-rated health) characteristics on the identified profiles. The traditional-reciprocity group served as the reference category in the regression model. Model fit was assessed using several indices including the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), sample-corrected BIC (aBIC), entropy, Lo-Mendell-Rubin (LMR) and bootstrap-based likelihood ratio test (BLRT).

The LPA identified five distinct profiles of non-coresidential intergenerational relationships: distant-emotional (high normative solidarity, high associational solidarity, low structural solidarity), proximity-detached (low scores across all dimensions), proximity-loose (high structural solidarity, moderate normative solidarity, low scores on other dimensions), distant-support (high functional solidarity despite low structural solidarity), and traditional-reciprocal (high scores across all dimensions). The proximity-loose profile was the most prevalent (40.16%). Multinomial logistic regression revealed significant effects of several variables on the likelihood of belonging to each profile. Males were more likely to be in the proximity-detached profile than females. Rural residents were more likely to be in the distant-support profile than urban residents. Individuals without legal spouses were more likely to be in the distant-emotional and distant-support profiles. Lower education levels were associated with higher likelihoods of being in the distant-emotional, proximity-detached, and proximity-loose profiles. Individuals with lower incomes were less likely to be in the traditional-reciprocal profile. Self-rated unhappiness and average happiness were associated with higher likelihoods of being in certain profiles other than the traditional-reciprocal group, indicating a strong correlation between self-reported well-being and the type of intergenerational relationship.

The findings highlight the significant diversity of non-coresidential intergenerational relationships in China. The prevalence of the proximity-loose profile suggests a weakening of traditional filial support despite geographical proximity. The influence of socio-demographic factors underscores the need for nuanced social welfare policies. The association between self-rated well-being and relationship profiles emphasizes the importance of considering the mental health of both older adults and adult children. The study challenges some assumptions from prior research, such as the simple link between income and support type. This study provides a more comprehensive understanding of intergenerational support beyond previous research, indicating that it is not merely economic and material exchanges but emotional, normative, and associational dimensions that dictate the types of relationships generated.

This study offers valuable insights into the complexity of non-coresidential intergenerational relationships in China. It emphasizes the need for policy interventions that address the diverse needs of both older adults and adult children. Future research should explore the longitudinal dynamics of these relationships and further investigate the role of consensual solidarity and potential intergenerational conflict. The study also suggests further investigation of the impact of macro-social security systems on these relationships.

The study's cross-sectional design limits causal inferences. The exclusion of the consensual solidarity dimension due to data limitations restricts a full understanding of value alignment within families. Future studies should address these limitations through longitudinal data and a more comprehensive measurement of intergenerational solidarity. The reliance on self-reported data may introduce some bias. Additionally, the study primarily focuses on adult children's perspectives, while incorporating the older adults' perspectives could provide a more complete picture.